ACL91 in Practice en

ACL IN PRACTICE
ACL IN PRACTICE
Copyright 2007 ACL Services Ltd.

All rights reserved.
No part of these materials may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means (photocopying, electronic, mechanical, recording, or otherwise), without permission in writing from the publisher, except by a reviewer
who may quote brief passages in a review.
ACL 9
December 2007
ACL Services Ltd.
1550 Alberni Street
Vancouver, BC
Canada V6G 1A5
Telephone: +1-604-669-4225
E-mail: info@acl.com
Web: www.acl.com
Printed in Canada
ACL, the ACL logo, the ACL logo with the text, ACL Data you can trust. Results you can see. and Audit Command Language are
trademarks or registered trademarks of ACL Services Ltd.
Microsoft, Windows and Windows Server are trademarks or registered trademarks of Microsoft Corporation. z/OS is a registered
trademark of IBM Corporation. SAP, R/2 and R/3 are trademarks or registered trademarks of SAP AG. Crystal Reports is a
trademark or registered trademark of Business Objects SA. All other trademarks are the property of their respective owners.
200711071500
C ONTENTS
Chapter 1: Introducing Metaphor Corporation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Tutorial objectives .............................................................................................................. 2
About Metaphor Corporation ............................................................................................ 3
Metaphor credit card policies ..................................................................................... 4
Acceptable category codes .......................................................................................... 5
Metaphor data files ...................................................................................................... 5
Plan your analysis ............................................................................................................... 6
Preliminary investigation ............................................................................................ 6
General tips to consider ............................................................................................... 7
Chapter review .................................................................................................................... 8
Chapter 2: Examine employee data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Examine the employee list ................................................................................................ 10
Open and view the employee data project ................................................................ 10
Get a statistical picture .............................................................................................. 11
Look for duplicate records ........................................................................................ 13
Examine employee salaries and bonuses ......................................................................... 14
Find total bonuses paid in 2002 ................................................................................ 15
View salaries in descending and ascending order .................................................... 15
View employees hired in 2002 ................................................................................... 16
Determine bonus as a percentage of salary ............................................................... 17
Exercises ..................................................................................................................... 18
Chapter review .................................................................................................................. 19
Chapter 3: Set up your project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Start your project in ACL ................................................................................................. 22
Create a project in ACL ............................................................................................. 22
Import your data files as tables .................................................................................. 23
Exercise ...................................................................................................................... 27
iii
Check and verify the data ................................................................................................

Check the data type and format ...............................................................................
Verify the data in a table ...........................................................................................
Document your work ...............................................................................................
Add record notes to a table .......................................................................................
Exercise .....................................................................................................................
Chapter review .................................................................................................................
27
28
30
30
31
32
32
Chapter 4: Begin Your Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Determine the total liability .............................................................................................
Extract expiring card numbers ........................................................................................
Find transactions by customer number ..........................................................................
Check the credit card balances ........................................................................................
Exercises ...........................................................................................................................
Chapter review .................................................................................................................
33
34
34
35
37
39
39
Chapter 5: Examine expense patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Find the total expenses for each category code ...............................................................
Summarize the total expenses for each category .....................................................
Isolate transactions by category ...............................................................................
Isolate particular recreational expenses ...................................................................
Exercises ....................................................................................................................
List the amounts for each card number ...........................................................................
Exercise .....................................................................................................................
Chapter review .................................................................................................................
41
42
42
43
45
46
46
47
48
Chapter 6: Analyze transactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Determine which transactions are unacceptable ............................................................
About the Join and Relations commands ................................................................
Check the code tables for obvious errors .................................................................
Identify unacceptable transactions ..........................................................................
49
50
50
51
53
iv
Identify the employees ...................................................................................................... 56

Identify the employees departments ........................................................................ 57
Create a list of employees .......................................................................................... 59
Report your findings ................................................................................................. 61
Exercises ..................................................................................................................... 62
Chapter review .................................................................................................................. 62
Chapter 7: Validate, correct, and extract data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Import the May transaction data ...................................................................................... 64
Use folders to organize your project ......................................................................... 64
Import transaction data for May 2003 ...................................................................... 65
Compare and correct the table layouts ..................................................................... 66
Combine and verify the tables .......................................................................................... 67
Combine the May transaction tables ........................................................................ 67
Verify the new table ................................................................................................... 68
Check the combined table for duplicates .................................................................. 69
Exercises ..................................................................................................................... 69
Isolate and correct validity errors ..................................................................................... 70
Check the format of card numbers ........................................................................... 70
Create a 16-digit card number field .......................................................................... 71
Extract the corrected records to a new table ............................................................. 72
Export the combined transaction table to Excel ....................................................... 73
Create an error-free employee record table .............................................................. 73
Exercises ..................................................................................................................... 74
Ideas for further analysis .................................................................................................. 74
Chapter review .................................................................................................................. 75
vi
C HAPTER 1
I NTRODUCING M ETAPHOR C ORPORATION
Experiment with real-world data analysis scenarios
In this chapter
Review the tutorial objectives
About Metaphor Corporation
Look at your files
Plan your project
ACL in Practice is a tutorial that examines human resources records and

credit card transactions at a fictional company, Metaphor Corporation. As
you work through the procedures and examples, you learn to apply ACLs
analysis and reporting capabilities to real-world data. When you finish, you
will be able to apply the power of ACL to your own data.
Chapter 1: Introducing Metaphor Corporation
Tutorial objectives
In this tutorial you will:
Work with an existing data analysis project
Create and work with a new ACL project from start

to finish
Report results of your ACL project
Export the results to another file format
These steps are typical for many data analysis projects

that you perform in ACL, both small and large.
Chapter 2: Examine employee data
Chapter 4: Begin Your Analysis

Objective: Determine basic background information
about the credit card numbers and transactions.
Summary:
Determine the total credit limit exposure of the

company.
Determine which credit card numbers are due to

expire within a given time period, and extract this
information.
Find which transactions are linked to a particular

customer.
Determine some basic information about the

credit card balances.
Objective: Examine employee profile data for Metaphor

Corporation.
Summary:
Examine the project and its contents
Create a statistical overview
Examine employee bonuses and salaries
Chapter 5: Examine expense patterns

Objective: Find information about itemization of
expenses and spending patterns.
Summary:
Determine transactions per merchant category

code.
Separate the transactions for airfare, hotel stays, or

rental cars.
Determine expenses for drinking establishments

and golf courses.
Determine total transaction amounts per credit

card number.
Determine transaction amounts per customer

code.
Chapter 3: Set up your project

Objective: Create the project and add the data.
Summary:
Create an ACL project.
Import the data into your project. Each data

source is a separate table in the ACL project.
Verify the data.
Document your work.
ACL in Practice
Chapter 6: Analyze transactions

Objective: Determine whether transactions are
acceptable, unacceptable, or require approval.
Summary:
Determine which transactions are unacceptable,

and which employees made these transactions.
Create a report that includes employees with

unacceptable transactions, details of the
transactions, and the employees departments.
Chapter 7: Validate, correct, and extract data

Since 1995, Metaphor has sold software and IT services.
Metaphor has several dozen consultants who are full-time
employees. These consultants work out of Metaphors
head office and they also travel the world in order to
install software, provide training courses, and initiate and
maintain sales relationships.
Metaphor also has administrative and professional staff
members in the head office. In addition, Metaphor has
sales agents in various locations around the world who are
not employees.
Objective: Expand your analysis to include more data

files.
Summary: This portion of the tutorial extends the
Metaphor Corporation scenario.
Import the May 2003 credit card transaction files

as tables.
Clean up the data in the May 2003 credit card

transaction tables.
Extract the repaired data to new tables.
Combine the two May 2003 credit card transaction

tables into one.
Using the techniques you have learned, analyze the

May 2003 credit card transactions.
ACL in Practice
Metaphor credit card policies

Most employees have a company credit card. When an
employee purchases products or services for the
company, he or she must use a company credit card.
Employees entertain clients and prospective clients as
appropriate, both in Metaphors home city and while
travelling. They are also responsible for their own
professional development, and are expected to charge
expenses to the company credit card for courses, books,
journal subscriptions, and professional memberships.
Some employees are also responsible for purchasing
supplies for the office.
Thus, Metaphor Corporation incurs significant expenses
in several categories:
By using ACL to analyze the credit card transactions of

Metaphor employees, you can find transactions that are
contrary to policy and transactions that might be
erroneous or fraudulent.
The following table shows allowable expenses.
Travel
Entertainment
Office and
professional
Airfare
Restaurants
Hotel
Drinking places
Dues for
membership clubs
Taxi
Boat
charters/rentals
Restaurants
Rental cars
Long-distance bus
tickets
Caterers
Bakeries
Dues for
professional
organizations
Dues for fraternal
organizations
Travel
Entertainment
Client service
Professional development
Amusement parks
Computers,
computer parts, and
software
Internal office expenses
Golf courses
Office supplies
Metaphors policy is that all purchases must be assigned to

a customer number. Although most of the customer
numbers correspond to external clients, customer
numbers that start with zero correspond to internal
departments.
Travel agencies
Toll and bridge fees
Sporting events
Recreational
services
Books and
periodicals
Tourist attractions
Theater tickets
Tour buses
Musicians
Bowling alleys
Employees are aware that certain categories of

transactions are unacceptable. For example, employees
must not use corporate credit cards for gambling.
Although the company sometimes approves transactions
in other categories, permission must be granted in
advance. Some categories are prohibited entirely.
ACL in Practice
Acceptable category codes

The credit card analysis in this tutorial relies on the
Merchant Category Code list, which is used by major
credit card providers. It contains thousands of four-digit
numeric codes, including unique codes for many major
airlines and hotel chains, and numerous generic codes.
Every credit card transaction has an accompanying
category code assigned to it.
In the real world, credit card companies occasionally
assign incorrect codes, so if you obtain questionable
results, contact the merchant for verification.
For the purposes of this analysis, you can compare
transaction codes to the lists of acceptable and
unacceptable codes. For example, if a Metaphor employee
uses his credit card to gamble, the transaction code
indicates an unacceptable transaction.
The Policies.doc document, which accompanies the
tutorial data, discusses appropriate credit card usage
policies. The data analyst uses this document to
determine what types of transactions to flag.
ACL in Practice
Metaphor data files

When you perform an analysis for your own company,
you often obtain the data in whatever file format is
available. In this tutorial, however, you continue an
analysis project that has already been planned. Much of
the preliminary work is already complete.
For this analysis project, you use these files:
Acceptable_Codes.mdb
This Access database file has a list of acceptable
codes from the Merchant Category Code list.
Credit_Cards_Metaphor.xls
This Excel worksheet lists credit card information.
Company_Departments.txt
This tab-delimited file contains a list of
departments and their internal numbers.
Employees.cvs
This comma-delimited text file lists the employee
names, employee numbers, and credit card
numbers.
Trans_April.xls
This Excel worksheet lists the credit card
transactions for April 2003.
Unacceptable_Codes.txt
This tab-delimited text file contains a list of
unacceptable codes from the Merchant Category
Code list.
Plan your analysis

Developing an analytical strategy is typically an iterative
process. As you work with the data in ACL, some patterns
will likely become apparent, which might lead to new
strategies. Here is a typical cycle:
Plan the analysis
Analyze the data
Reconsider the analysis
Analyze the data again
As you develop and work through your analysis, you

might decide that you need additional data. For example,
you use hire date and salary information in your analysis,
but once you assemble preliminary results, you decide to
acquire employee absence records. You can always add
more data to your project, but obtain as much raw data as
you can before you begin your analysis.
An ACL analysis involves a project that contains one or
more tables. These tables are imported from external data
sources, such as Excel worksheets, Access databases, text
files, Oracle databases, and so forth. In a typical
corporation, various departments store and assemble
data on a daily basis. You can use ACL to combine data
from disparate sources to create a single coherent analysis.
Once you have your data in place, you can develop
objectives for your analysis. Consider some possible
approaches. Make a note of tests that you think might be
useful.
Preliminary investigation
Because the tutorial includes ready-made data, you do not
have to perform any preliminary investigation of the data.
In general, however, you should investigate the data you
want to use before you proceed.
Consider these questions:
What data currently exists, and what format is it

in?
Who entered the data?
Who maintains the data?
Which department owns this data?
Is the data accurate and properly formatted? If not,

can the data be modified to make it suitable?
Can you combine the data from more than one

source?
Can you derive the data you need from the data
you already have?
Before you import a file into your project, look at the file
in its original format. For example, you can open a
Microsoft Excel worksheet to see what it looks like, or you
can open a text file in Notepad.
In a large file, such as a database that contains millions of
records, it might be impractical to look at the data in its
original format. In that case, you can ask your IS
department for a description of the fields and the field
sizes. If possible, review an excerpt of the file to get a
better understanding of its content.
For example, if you want to analyze some aspects of
customer invoices over the period 1999 to 2002, the data
might exist in a comprehensive database that is in a form
ACL in Practice
Plan your analysis
that is not easily accessible. You could ask your IS

department for information about specific fields, which
might look like this:
Field
Comments
Company Name
30 characters, Text
Amount
12 characters, Text
Date of Invoice
10 characters, Date,
DD/MM/YYYY
Date_Payment
10 characters, Date,
MM/DD/YYYY
With this type of information, you know in advance how

the data appears. For example, you know that the two date
fields are formatted differently.
Although ACL is capable of handling difficult data, you
can make the process easier for yourself by ensuring that
your data is clean.
ACL in Practice
General tips to consider

Consider the following steps when you begin to organize
a new project.
Refine your strategy with test data. There is little

need to restrict yourself to using data samples
because ACL can read data files of any size.
However, small extracts of data might be useful
when you are developing an analytical strategy.
For example, you can look at at one days worth of
data before you analyze five years worth. Working
with a thousand-record file is much quicker than
working with a million-record file. When you have
your strategy in place, you can apply it to a
million-record file.
Eliminate errors from your data. The quality of

your analysis will improve with the quality of your
data. Try to eliminate any errors from your data
before bringing it into ACL. If that is not possible,
you can use ACL to correct the data.
Experiment with your data. If you arent sure of

what an ACL function or command will do, you
can try it anyway. Your data remains unharmed.
Chapter review
In this chapter you learned:
The tutorial objectives.
The Metaphor credit card policy.
How to plan a typical ACL project.
ACL in Practice
C HAPTER 2
E XAMINE EMPLOYEE DATA
Analyze salaries, bonuses, and hiring dates
In this chapter
Examine the employee list
Examine salaries and bonuses
In this chapter, you look at some Metaphor Corporation employee profile

data. You analyze salaries, bonuses, hiring dates, and other human resources
information. In the course of this analysis, you learn the basics of ACL,
practice navigating the program, and analyze data.
In this scenario, you take over an ACL project from a coworker who has
already started work on it. You analyze the data that your coworker
assembled.

When you look at the employee data, you can get some
ideas about how you might analyze this data. In this
example, the project has already been assembled for you.
When you work with your own projects, you decide what
data to include. You can also add more data at any time to
an existing project.
Your task is to determine various financial and human
resource information, based on the data in your ACL
project.
The project you work with in this chapter represents a
small-scale, but fairly typical data analysis project. This
project has two tables that contain information about
Metaphor employees. You perform the analysis based on
this information.
As you work through the tutorial, you might notice that the
columns on your screen sometimes appear in a different order
from the columns in the illustrations. To change the order of a
column, click the field list heading in the dialog box with which
you are working.
Tasks
View the employee data project.
Get a statistical picture.
Check for duplicate employee records.
10
Open and view the employee data project

You begin your examination of the employee data by
viewing the contents of the project.
An ACL project is a holder for data that you import into
ACL as tables. Once the data is imported into the project,
it does not matter to ACL what the original data source
was.
When you use ACL, you either create a new project to
work with, or you open an existing project. You open,
save, and close projects in ACL in much the same way that
you manage documents in other Windows programs.
Open the employee data project
1 Open ACL.
2 Select File Open Project. Browse to locate the
C:\ACL Data\Sample Data Files folder. Select
Metaphor_Employee_Data. Click Open.
If your installation of ACL did not use the default folders, this file
may be located elsewhere on your system.
Notice that the Project Navigator pane is at the left of the
window, and the table view is at the right. On the
Overview tab, you can access the tables, scripts, and log.
In this project there are two tables: Agents_Metaphor
and Employee_List. The Agents_Metaphor table lists all
of Metaphors independent sales agents around the world.
The Employee_List lists all of the Metaphor employees at
the head office.
ACL in Practice
View the contents of the employee data project
1 In the Overview, right-click

Metaphor_Employee_Data.ACL and select
Properties. Click the Notes tab.
Although this project contains no notes, you can add
them at any time to record details of your work.
2 Close the Project Properties dialog box.

3 Double-click Agents_Metaphor in the Overview to
open the table.
Get a statistical picture

The Statistics and Profile commands give you a quick
statistical snapshot. With larger tables, the information
that you can get from Statistics and Profile is often useful.
The Statistics command works with numeric and date
fields. Profile works with numeric fields only.
In this procedure, you generate statistics about the
Employee_List table, such as ranges of hiring dates and
salaries.
4 Select Analyze Count Records. The Count dialog

box appears. Click OK.
The count of 41 records appears at the bottom of your
screen. Although you cannot see a difference on the
screen, ACL wrote the information to the log.
Generate statistics on the employee list
1 With the the Employee_List table active, select

Analyze Statistical Statistics. The Statistics
dialog box appears.
5 Open the Employee_List table. Count the number of

employees. How many employees are there? You
should get 200.
Before you run a command in ACL, ensure that no column in

the view is selected. Otherwise, ACL will run the command on
just the selected field.
ACL in Practice
11
2 Click the Statistics On button. The Selected Fields

dialog box appears. Click Add All.
If you prefer, you can run the Statistics command on each

field separately, so that you see the results separately.
The Profile command also gives you an overview of the
data that is more concise than that provided by Statistics.
Use Profile to get information about a table
1 With the Employee_List table active, select Analyze

Statistical Profile. The Profile dialog box appears.
3 Click OK.On the Output tab, ensure that Screen is

selected, Click OK. ACL displays the results.
2 Select all of the fields and click OK. ACL displays the
results.
If you scroll down, you can see that the earliest hiring date
is 01/05/1995 and the latest hiring date is 12/19/2002.
Some of the information provided is not useful, such as
the total of the HireDate field values, so you can disregard
this information.
12
Each of the values for each of the numeric table columns is

shown on the tab.
ACL in Practice
Look for duplicate records

One way to ensure the integrity of the data is to check for
duplicate records.
This might be a more complicated task than it seems at
first glance. One employee might be listed under two
names, or under the same name twice. Two employees
might have been assigned the same employee number
inadvertently. The type of duplicates you want to check
depends on the data that you are working with.
Although you are checking for duplicates in only one

field, include all of the fields in the output screen to
help you to interpret the results.
4 On the Output tab, select Screen. Click OK.

ACL displays the results.
In the next procedures, you check for duplicates on the

employee last name and the employee number.
Look for duplicate last names
1 With the Employee_List table active, select Analyze

Look for Duplicates. The Duplicates dialog box
appears.
5 In the Last_Name column, click a hyperlink. ACL

displays a filtered view of the Employee_List,
showing that employees details.
You next decide to check the Employee_List to determine
whether there are any duplicate employee numbers.
Look for duplicate employee numbers
1 Click the Remove Filter

2 Click Duplicates On. The Selected Fields dialog box
appears. Click Last_Name and click the arrow button.
Click OK.
3 Click the List Fields button. The Selected Fields

dialog box appears. Click the Add All button. Click
OK.
ACL in Practice
button to remove the filter.

The full Employee_List table appears again.
2 Select Analyze Look for Duplicates. The

Duplicates dialog box appears.
3 Click Duplicates On. The Selected Fields dialog box

appears. Select EmpNo and click the arrow button.
Click OK.
13

5 View the Duplicates tab. No duplicate employee
numbers are found.
Examine employee salaries and

bonuses
You have been asked to find several things from the
employee data:
The total bonuses that were paid out in 2002.
The salaries paid: from highest to lowest, and

lowest to highest.
The number of people who were hired in 2002.
The ratio of bonus to salary for each employee in

2002.
Tasks
Find the total amount of paid bonuses.
View salaries in ascending and descending order.
View employees hired in 2002.
Calculate bonus as a percentage of salary.
14
ACL in Practice
Examine employee salaries and bonuses
Find total bonuses paid in 2002

Metaphor has a complex compensation structure.
Because Metaphor is a project-oriented company, some
employees have a compensation scheme that relies heavily
on bonuses.
To find the total amount paid in bonuses in 2002, use the
Total command.
Use the Total command to find total bonuses paid in 2002
1 Open the Employee_List table.

2 Select Analyze Total Fields.
3 Click the Total Fields button. The Selected Fields
dialog box appears. Select Bonus_2002 and click the
arrow button. Bonus_2002 now appears in the
Selected Fields list. Click OK. Click OK.
4 View the Total tab.

The figure shown on the Total tab is the total amount
paid in bonuses in the year 2002. The total that
appears should be 1,016,483.76.
Suppose you wanted to see what command you just
entered to achieve this result. You can look at the log.
View salaries in descending and ascending order

The Quick Sort feature lets you view the data in sequential
order. Applicable only to the view, Quick Sort is a
temporary means to view your records in ascending or
descending order. When you are finished with Quick
Sort, you can turn it off to return the view to the order in
which the records exist in the table.
View employee salaries in descending and ascending order
1 With the Employee_List table active, right-click the

Salary column header. Select Quick Sort
Descending.
You can now see the employee salaries in descending
order. You decide you would like to see them in
ascending order.
2 Right-click the Salary column header. Select Quick

Sort Ascending.
If you have a large file, Quick Sort operations can be
lengthy. If you have small files, you might find Quick Sort
useful. You can Quick Sort on any column, so you can get
different perspectives on the data.
5 Click the Log tab at the bottom left of the ACL

window.
You can see the commands that you performed on this
project.
A command is the name of a specific action taken in ACL, such as
the Statistics command. To learn more about commands, select
Help Index and look up commands.
ACL in Practice
15
View employees hired in 2002

To determine how many employees were hired in 2002,
you apply a filter to the view. A filter is a logical expression
that you can type in the views Filter text box. ACL
displays only those records that match the filter.
An expression is a set of operators and values that you use to
perform calculations, specify conditions for a test, or create values
that do not exist directly in the data. An ACL expression can be a
combination of data fields or computed fields, operators,
constants, functions, and variables.
5 Change the expression to read:

HireDate >= `20020101` AND HireDate <=
`20021231`
6 Click OK.
7 View the filtered records.
8 Select Analyze Count Records and click OK.
Look at the bottom of the ACL window. Note that there
are 17 records showing up with this filter. This means 17
employees were hired in 2002, according to this list.
Create a filter to show employees that were hired in 2002
1 Click the Edit View Filter
button. The Edit View

Filter dialog box appears. You can use this dialog box
to create and edit filters that apply to the current table.
2 In the Available Fields list, double-click HireDate.

3 Click >= in the middle pane.
4 Click Date. Locate January 1, 2002 and click OK.
The expression should read:
HireDate >= `20020101`
Note that any filter that involves dates uses reverse

single quotation marks. You cannot use typical
quotation marks when you use dates, and you cannot
use this special quote mark in other expressions.
Although you know that no employees have been
hired in 2003, you want to ensure that the time period
that you selected is correct. You want a list of all
employees hired between January 1, 2002 and
December 31, 2002.
16
ACL in Practice
Examine employee salaries and bonuses
Determine bonus as a percentage of salary

You want to determine the bonus as a percentage of salary
for all employees. To do this, you add a computed field to
the table layout.
A computed field is created with an expression. This
expression can use one or more fields already in the table.
6 In the Expression box, enter the expression:

(BONUS_2002 * 100.00) / Salary
This expression determines the 2002 bonus as a

percentage of the employees 2002 salary.
7 Click OK.
8 In the Format list, enter 999999.99%.
Add a computed field to the table layout
1 With the Employee_List table active, click the

Remove Filter
2 Select Edit Table Layout. The Table Layout window

opens with the Edit Fields/Expressions tab
displayed.
To learn more about how ACL handles precision and accuracy

in decimal calculations, select Help Index and look up
controlling rounding in decimal expressions.
9 To save the changes, click the green checkmark.

10 Close the Table Layout window.
Although you have created the new field, it does not
yet appear in the Employee_List view.
11 Right-click the CardNum column header and select

Add Columns. The Add Columns dialog box
appears.
12 Select Bonus_Percent_Salary and click the arrow

3 Click the Add a New Expression
button on the left.

The Edit Fields/Expressions tab changes so that you
can enter the expression for the new computed field.
4 Enter Bonus_Percent_Salary in the Name field.

5 Click f(x). The Expression Builder dialog box
appears. The Expression Builder is the same as the
Edit View Filter dialog box that you used already.
ACL in Practice
button. Bonus_Percent_Salary now appears in the

Selected Fields list. Click OK.
Keep in mind that the ratio will be skewed for employees hired in
2002, because they did not work a full year at Metaphor. You can
apply a filter to hide the records of employees hired in 2002.
17
When working in an ACL project, you might want to

resize the widths of the columns. You can move the
column separator to adjust the width of the column so
that it occupies less space in the view. Alternatively,
you can double-click the right side of the column
heading to resize the column to the width of its visible
display. To resize more than one column at a time,
select multiple columns, then double-click the right
side of a column heading.
13 Close the view and click Yes when ACL prompts you
to save.
Exercises
1 Can you run the Profile command on just the
HireDate in the Employee_List? Run the Statistics
command on just the HireDate.
What results do you get? When are the Statistics and
Profile commands useful?
2 Quick Sort the Employee_List on the Bonus_2002

column.
3 Quick Sort Agents_Metaphor on the Comm_2002

column in ascending order, then in descending order.
Run Statistics and Profile on this table.
4 Find the total commissions paid to Metaphor agents

in 2002.
5 Make the view easier to read by adjusting column

widths and dragging columns left or right. You can
adjust the columns at any time while you follow the
tutorial.
18
ACL in Practice
Chapter review
Chapter review
In this chapter you accomplished the following tasks:
Opening a project
Viewing the project contents
Using Statistics and Profile commands to analyze

data
Using Quick Sort to sort the data in a specified

order
Creating and applying filters
You have completed a small demonstration of what you

can do with ACL. You work through more complex
examples in subsequent chapters.
ACL in Practice
19
20
ACL in Practice
C HAPTER 3
S ET UP YOUR PROJECT
Create the project and add the data
In this chapter
Manage projects in ACL
Import files into a project
Check and verify the data
Before you begin your analysis of Metaphor Corporations credit card

transactions, you must create a project and access the credit card data.
An ACL project functions as a container for data. You import the source data
into your project from existing data sources, such as Excel worksheets, text
files, Access databases, dBASE files, and so forth. You import each data
source as a separate table. Once you import the data, you can then analyze it.
21
Start your project in ACL

Because you are starting a new data analysis project, you
begin by creating a project in ACL.
Create a project in ACL

To begin your work in ACL, create a project. You can save
a project or close an ACL project just as you do with a
word processing document or a spreadsheet.
Tasks
Create a project in ACL.
Import your data files as tables.
Other data import scenarios.
Create a project in ACL
1 Select File New Project.

2 In the File name box, enter
ACL_Tutorial_Metaphor and click Save.
The Data Definition Wizard appears. You use this

wizard to import data into your project.
If you need to change a previous selection when you use the Data
Definition Wizard, click Back until you return to the screen that
you want.
22
ACL in Practice
Import your data files as tables

Now that you have begun a project in ACL, you add data
to it. In each instance, you use the Data Definition
Wizard, but the process varies slightly, depending on the
type of file that you import.
You import data from each of the six files listed under
Metaphor data files on page 5. ACL creates and saves a
table for each file in a .fil file.
For the first file that you import, the Data Definition
Wizard is already open because it appeared when you
saved the new project. Start by importing the
Credit_Cards_Metaphor.xls file, which ACL recognizes
as an Excel file.
Import the Credit_Cards_Metaphor.xls worksheet
1 On the Data Definition Wizard screen, click Next

and proceed through the wizard until the Select File
to Define dialog box appears. Browse to find the
C:\ACL Data\Sample Data Files folder. If your
installation did not use the default locations, this
folder might be located elsewhere on your system.
2 Double-click Credit_Cards_Metaphor.xls.
3 Click Next. ACL displays the Data Source screen.
4 Select the worksheet. Enter a value in the Maximum
Character Field Length box. Click Next. ACL
displays the Save Data File As dialog box.
5 Enter Credit_Cards_Metaphor. Click Save.

You get a listing of the fields with a description.
6 Click Finish. When a message appears that asks you to

save the untitled table, click OK.
In this example, the Excel workbook

Credit_Cards_Metaphor.xls contains only one
worksheet. If an Excel workbook has more than one
worksheet, you select which worksheet to import on the
Data Source screen of the Data Definition Wizard.
Import the Trans_April.xls worksheet
1 Select File New Table. The Data Definition

Wizard appears. Click Next until the Select File to
Define dialog box appears.
2 Locate and double-click Trans_April.xls. Click Next.

3 Proceed through the wizard as you did with the
previous table. Save this table as Trans_April.
4 Click Finish. Click OK.

The procedures for importing the other data files as tables
are similar to the procedure for importing Excel
worksheets.
Import the Unacceptable_Codes.txt file
Unacceptable_Codes.txt has two columns titled Codes
and Description. The field entries are separated by tab
characters.
It is best to give each field the correct data type when you import
it into ACL. However, you can change the data type after you
import the file.
1 Select File New Table. Click Next until the Select

File to Define dialog box appears.
2 Locate and double-click Unacceptable_Codes.txt.

Click Next.
The Credit_Cards_Metaphor table opens in ACL.
ACL in Practice
23
3 On the File Format screen of the Data Definition

Wizard, select Delimited text file, and click Next.
The Delimited File Properties screen of the Data
Definition Wizard appears.
4 Select TAB in the Field Separator column. The data

separates correctly into two fields.
5 Select Use first row as field names.

The first row of the file contains the field names, which
are Codes and Description in this instance. You can
see the format of the file as defined in the wizard.
Although the unacceptable codes are comprised of

digits, you want ACL to treat these codes as character
data, rather than numeric data.
8 Click the Description column header on the Edit

Field Properties screen. It should have a Type of
ASCII Text. You do not have to make any changes.
9 Click Next.
The wizard provides a summary of the file.
10 Click Finish. Click OK when you are prompted to

save the table.
The Unacceptable_Codes table appears in ACL. You can
scroll through the contents.
Import the Company_Departments.txt file
This file is similar in format to the
Unacceptable_Codes.txt file, but it has no header row.
You must add the field names in the Data Definition
Wizard.
1 In the overview select the
6 Click Next and save the table as

Unacceptable_Codes.
The Edit Field Properties screen of the wizard

appears. You can change the type of the data in each
field. ACL has made assumptions about the data that
are not necessarily correct.
7 On the Edit Field Properties screen of the wizard, the

Codes field is selected by default. Select ASCII Text
from the Type drop-down list.
24
ACL_Tutorial_Metaphor project folder and

right-click and select New Table to open the Data
Definition Wizard. Proceed through the wizard as
you did for the previous file, but this time select
Company_Departments.txt.
2 Again, select Delimited text file on the File Format

screen. Click Next.
3 On the Delimited File Properties screen, deselect

Use first row as field names if necessary. Select TAB
as the Field Separator.
4 Click Next. The Save Data File As dialog box appears.

5 Enter Company_Dept as the File name. Click Save.
ACL in Practice
You can now edit the field properties.
6 On the Edit Field Properties screen, enter

Dept_Name in the Name text box.
7 Click the Field 2 column header. Enter Dept_Code

in the Name field. From the Type drop-down list,
select ASCII Text.
8 Click Next. Click Finish. Click OK to the message.

The Company_Dept table opens in the view.
Import the Employees.csv file
1 Select Data External Data Disk. The Select File to

Define window appears.
2 Double-click the Employees.csv file to open the Data

Definition Wizard.
3 On the File Format screen, ensure that ACL imports

the file as a Delimited text file. Click Next.
4 On the Delimited File Properties screen, ensure that:
Comma is selected as the Field Separator.
Use first row as field names is selected.
ACL in Practice
5 Click Next. Save the file as Employees.

6 On the Edit Field Properties screen:
Change the CardNum field and the EmpNo field

to ASCII Text type.
Change the HireDate field to Date type. Specify

the Date Format. If the file you define does not
match the date format you will get blank dates in
the view.
Ensure that Salary and Bonus_2002 are Numeric

(Formatted) type.
You can change the column titles if you want more

user-friendly titles. Changing the titles does not
affect the field names.
7 Click Next. Save the table as Employees.

8 Click Finish. Click OK to the message.
The Employees table opens in the view.
You next import the data from the Acceptable_Codes
database.
Import the Acceptable_Codes Access database
1 Click File New Table. The Data Definition

Wizard appears.
If the welcome message also appears, select Do not
show this welcome screen again, and click Next.
2 Select the Acceptable_Codes.mdb file.

3 Continue to proceed through the wizard. Since this
database includes only one table, Acceptable_Codes
is automatically selected in the Select the Access
Table/View list.
4 Save the table as Acceptable_Codes.
25
5 Click Finish. Click OK to the message. The

Acceptable_Codes table opens in the view.
As you can see in the Overview at the left of your project,
the data from each file has been placed into separate
tables. You should have six tables in your project:
Acceptable_Codes
Company_Dept
Credit_Cards_Metaphor
Employees
Trans_April
Unacceptable_Codes
table, or when you close the project. Click Yes to save the
work you have done on each table.
You can stop following the tutorial at any point by closing and
saving the project.
Save your project
Click File Save Project. You can also click the

Save the Open Project button on the toolbar.
In order to work through this tutorial effectively, you must
save your work as you move from one procedure to the
next.
Just as with any Windows application, you should save
your work in ACL every few minutes while you are
working on a project. That way, you protect yourself from
losing your work if there is a system crash. ACL
automatically creates a backup copy of the working
project as well. You should save your work when you add
or remove data from your project.
Throughout the tutorial, you will be prompted to save
changes to a table, particularly as you move from table to
26
ACL in Practice
Exercise
Save the existing project under a new name,
ACL_Tutorial_Metaphor_Test, using File Save As.
Continue working with the original project,
ACL_Tutorial_Metaphor.

Before you proceed with analyzing your data, you should
always verify it. For example, if a field is supposed to
contain dates, but also contains names, your analysis may
be meaningless. ACL has several commands that you can
use to verify your data.
Tasks
Check the data type and format.
Verify the data in a table.
Document your work.
ACL in Practice
27
Check the data type and format
Names
The entries should contain only

letters, spaces, and hyphens. The
field should have a data type of
ASCII.
Merchant Category Codes
Each code should be 4 digits long,

with no extraneous characters. The
ASCII.
Before you begin your analysis, check the table layout for
each table to determine whether:
The data types for each field are correct.
The fields are properly defined.
The following table provides information about how the

fields in the tables should look, and what data type ACL
should assign to them.
The way the data should appear is a matter of preference.
For example, you could decide that you wanted all card
numbers to be formatted with hyphens after each four
digits. In this project, the decision was to display a string
sixteen characters long.
Files
Data type and formatting
Credit card numbers
Each entry should have 16 digits,

without any extra characters. The
ASCII.
Dates
Each entry should be a date.

Depending on the dates, you might
want to determine if all the dates
fall within a given time period. The
Date.
Transaction Amounts,
Finance Charges, and
Credit Limits
28
Each entry should be numeric, with

two decimal places. The field
should have a data type of Numeric.
In this tutorial, you assign the ASCII data type to

Merchant Category Codes and credit card numbers
because you will not perform calculations with either
field. You also require an ASCII data type for work that
you do later in the tutorial. If you were performing
mathematical tests on the card numbers, you would need
a data type of Numeric.
Ideally, in a thorough data analysis, you verify the data in
every field in every table of your project. If you like, you
can set up a chart with the appropriate tests for each field,
and keep a check list of the results. Each verification
command is also shown in the log.
However, if you are confident that your data has been
entered properly, you can skip the verification steps or
verify only a few fields.
Often, you can use ACL to fix incorrectly formatted data.
You can create computed fields that omit improper
characters, for example.
In this chapter, you will modify the default data types that
ACL selected, but the underlying data does not require
repair.
ACL in Practice
View and edit data definitions in a table.
1 Open the Credit_Cards_Metaphor table.

2 Click CTRL+I. The Table Layout window appears.
This window lists information about each of the data
fields as it is defined in the project. Look at Category
and Type to determine how the data has been defined.
You may need to scroll the list horizontally to see all of
the information. You can widen and rearrange
columns to make the list easier to read.
4
5
6
7
Double-click the CardNum entry.

In the Type drop-down list, select ASCII.
Click the green checkmark to save the changes.
Close the Table Layout window.
Field lists like this one categorize data types as Character,

Numeric, Date, or Logical. The ASCII data type, for example,
belongs to the Character category. To learn more about the
categories of data types, select Help Index and look up data
type reference.
3 Look through the list of fields to determine whether

each field has the correct data type.
The CardNum field should be character data, but the
N indicates that it is numeric data. Although it is
comprised of digits, you will not be performing
mathematical operations with the card numbers.
ACL in Practice
29
Verify the data in a table

Once you establish that the data types and the data are
correctly defined, you next determine whether the data
matches its assigned data type. Use the Verify command
to check whether the data in a table matches its data type.
For example, if you have letters in a field with a numeric
data type, or decimal numbers in a field with a date data
type, the Verify command indicates data validity errors.
Verify the data in a table

2 Click Data Verify.
3 Click Verify Fields. The Selected Fields dialog box
appears.
4 Click Add All. Click OK.

You should see a list of the fields you want to verify in
the Verify Fields list.
5 Click OK.
You should have zero data validity errors.
You have now verified all fields in the table.
Document your work

When you work with an ACL project, you want the
methods that you use in your work to be clear. You can
then justify the results of your work, or allow someone
else to take over the analysis tasks if necessary. With
proper documentation, you can see at a glance what work
you have done in ACL. This will make your job easier,
particularly if you are responsible for multiple analysis
projects.
If you are taking over a project from someone else,
documentation is particularly valuable. Similarly, if you
later assign the project to someone else, that person will
appreciate being able to read historical project notes.
Add and edit project notes
1 Select the project name in the Overview.

2 Click File Properties. The Project Properties
dialog box appears.
3 In the Notes tab, enter the current date and time, and
a short summary of what you plan to accomplish.
4 Click OK to close the Project Properties dialog box.

When you close the dialog box, ACL automatically saves
the notes. However, if you accidentally erase the notes,
you cannot get them back unless you have a file backup.
For this reason, it is recommended that you use both a
task sheet and the project notes for documentation.
If you find it helpful, you can adapt the following task
sheet to your project to keep track of what analysis you
have already done, and what analysis you plan to do. This
task sheet is prepared and maintained outside of ACL.
30
ACL in Practice
Add record notes to a table

Items
Information you need
Where to get this information
Project folder location
Related projects
Verification (create a checklist
for each verified field)
Table names with extracted,
validated data
Description of each analytical
task in ACL
Description
You can add notes to specific records in a table. These

notes can contain any information that might be helpful
to someone viewing the project, and you can also link to
other files containing information.
1 Open the Employees table. Note the numeric cells

that indicate the record number at the left of the table.
2 Right-click record cell 3 and select Edit Note. The

Edit Note dialog box appears.
3 Enter this note: This employee is leaving

Metaphor Corporation by the end of the
year.
4 Click OK.
Any notes you add to a table do not affect the table data.
Analytical task #1 to n... (ACL

steps)
Report for each objective
Reporting task #1 to n... (ACL
steps)
Location and description of
exported data files
ACL in Practice
31
Exercise
Verify the data for at least one other table in the
ACL_Tutorial_Metaphor project.
Chapter review
Creating a project.
Importing files into your project.
Saving your project.
Verifying the data in your project.
Documenting your work.
You have not actually done any analysis of the data yet, but
you have set the project up so that it is ready.
32
ACL in Practice
C HAPTER 4
B EGIN Y OUR A NALYSIS
Start looking for patterns
In this chapter
Determine total liability
Extract expiring card numbers
Find transactions for a customer
Check credit card balances
So far, you looked at a project plan, worked on a simple project, started a new
project, added data to the project, and verified the data. You now use ACL to
perform simple analysis of your data.
In this chapter, you use simple procedures to obtain general information
about your data. You look at the credit limits and the credit card balances to
get information about Metaphors financial liability and credit card usage
patterns.
33
Determine the total liability

If every Metaphor employee reached the credit limit on
his or her credit card, what would be the total debt for
Metaphor? You can determine this figure by totalling the
CREDLIM field.
You can use Total to add the total credit limits together to
determine the companys total credit exposure.
Total the CREDLIM field

2 Select Analyze Total Fields. The Total dialog box
appears.
3 Select the CREDLIM field. You only want to total this

field, so there is no need to select any other fields.
4 Click OK. View the Total tab.
Extract expiring card numbers

You plan to coordinate with the accounting department to
ensure that there is no problem with the employees
getting new credit cards. First, you filter all credit card
numbers that have an expiry date of January 1, 2004 or
earlier.
Filter card numbers by expiry date
1 On the Credit_Card_Metaphor tab, click the Edit

View Filter
box appears.
button. The Edit View Filter dialog
2 In the Expression box, enter this filter:

EXPDT <= `20040101`
If you prefer not to enter the expression, you can

double-click the field in the Available Fields list, click
the buttons for the expression symbols, and click the
Date button to select the date.
3 Click OK.
4 Select Analyze Count Records. Click OK. The
status bar shows that 8/200 records fit the criteria of
having an expiry date of January 1, 2004 or earlier.
What is your result? Your total should be 1,440,800.
You can export the results to another file format if you
want to save the result. You can also view the results in the
log if you come back to the project later.
34
ACL in Practice
Find transactions by customer number

You want to send the soon-to-expire credit card numbers
that appear in the view to the clerk in charge of credit
cards. In the next procedure, you export the credit card
numbers to a text file.
Export soon-to-expire credit card numbers
1 Click Data Export to Other Application. The

Export dialog box appears.
2 Select View. From the Export As drop-down list,

select Text.
Find transactions by customer

number
Because you have been warned about any expenditures
related to customer number 444413, you want to find
some information about the transactions associated with
that customer number.
You can accomplish this task in at least two different ways:
Create a filter using the FIND( ) function that

looks for all records with a CUSTNO value of
444413. You can use any of ACLs functions to
build a filter or computed field. Use FIND( ) to
isolate specific records.
Create a filter that sets CUSTNO equal to 444413.
3 Enter a descriptive filename in the To text box, such as

CCN_To_Expire_by_January1.txt. Click OK.
You can then attach the text file to an e-mail, print it out,
or manage it in whatever way you want.
You use functions to build either a filter or an expression for a

computed field. You can see the functions in the Expression
Builder.
Before you can do the search, however, check the data
types assigned to the fields in the Trans_April table.
Because you imported the data from Excel, ACL made
default assumptions about the data types for the fields.
You might need to change some of the data types.
Edit the Trans_April data types
1 Open the Trans_April table.

2 Click Edit Table Layout. The Table Layout window
appears.
ACL in Practice
35
Use the FIND( ) function to locate customer records
1 With the Trans_April table active, click the Edit View

Filter button. The Edit View Filter dialog box
appears.
2 In the Expression box, enter this expression:

FIND("444413",CUSTNO)
Quotation marks are used around 444413 because

the CUSTNO field has a field type of ASCII.
3 Click the Verify button. A message appears that

indicates that the expression is valid. Click OK. If the
expression is not valid, correct it and verify it again.
You do not have to verify the expressions you create,
but doing so is recommended.
The AMOUNT field has the correct data type of

NUMERIC.
The CARDNUM and CODES fields should be

edited to have a data type of ASCII.
4 Click OK. Scroll to the right to see that the CUSTNO
The CUSTNO and DESCRIPTION fields have the

correct data type of ASCII.
5 Select Analyze Count Records. Click OK. 26/281
The TRANS_DATE field has the correct data type

of DATE.
3 Double-click CARDNUM. From the Type
field only contains values of 444413.

records have a CUSTNO value of 444413. In a typical
analysis, you would likely try to find out who this
customer is. However, the data files that come with the
tutorial do not contain this information.
drop-down list, select ASCII. Click the green

checkmark. Repeat for CODES.

Now that you have checked and corrected the data types,
you can proceed to use the FIND( ) function.
You can also use FIND( ) to search for substrings within a

field. For example, the filter FIND("00",CUSTNO) would
36
ACL in Practice
Check the credit card balances

isolate all instances where 00 appears in the customer
number field, whether at the beginning, middle, or end.
Setting a filter directly
1 With the Trans_April table active, click the Remove

Filter

Filter dialog box appears. Enter the expression:
CUSTNO = "444413"
3 Click OK.
You get exactly the same results with this filter as you
did using the FIND( ) function.
4 Click the Remove Filter
When you apply a filter to data, and you want to save the
results, you have several options. You can:
Extract the filtered records to another table.
Create a report containing the filtered records.
Reapply the filter to the original data whenever

you want to look at it again.
As long as you have the original data, you can always

reapply the filter to see the filtered data again.
ACL typically provides multiple methods for performing
any given task. If you want to be sure that your reasoning
is correct, you can perform a task in more than one way
and check to see if you get the same result both times.
Check the credit card balances

You might get a general overview of your data to alert you
to possible areas of concern. For example, you might want
to find the highest balances, or find balances over a
certain limit. You can use this information as an indicator
of potential problems with employee credit card usage.
One way to determine how many credit card users have
high balances would be to filter the records to show those
with a NEWBAL value greater than 2000 and a NEWBAL
value greater than 1000. You can also sort the balances to
see the balances in order of size.
Filter the NEWBAL field to find balances that exceed $1000
or $2000
1 Open the Credit_Cards_Metaphor table. Click Yes if

ACL prompts you to save the changes to the
Trans_April table.
2 Right-click the NEWBAL column header. You may

need to scroll horizontally to locate the column. Select
Quick Sort Descending.
ACL lists the balances in descending order, with the
top value first. By scrolling through the table, you can
get an idea of what the various balances are.
Filter dialog box opens.
4 Enter this expression in the Expression box:

NEWBAL > 2000
5 Click OK.
ACL in Practice
37
The balances greater than $2000 are listed in

descending order.
10 Count the filtered records. You should have 89.

If you want, you can extract the filtered records to a new
table, or print a report.
You next decide to try another Quick Sort to see credit
cards with past due amounts.
Quick Sort on PASTDUEAMT
1 Click the Remove Filter button to remove the filter.

2 Right-click on PASTDUEAMT and select Quick Sort
Descending.
6 Select Analyze Count Records. Click OK.

The status bar shows 53/200 records that have a
NEWBAL amount greater than $2000.
Next, you change the filter to look at records with a
NEWBAL amount greater than $1000.
3 Scroll through the records. Notice that many credit

card holders have past due amount charges.
These procedures are minor examples of how ACL can
analyze a large amount of data. As you gain more facility
with ACL, you will undoubtedly think of new ways to
analyze your data.
7 In the Filter field at the top of the view, change the

filter to NEWBAL > 1000 and press ENTER.

You should have 65 records that fit the criteria.
You can have more than one condition in your filter. In
addition to the transaction amounts, the amounts in
the PASTDUEAMT and FINCHG fields might
indicate a problem with credit card usage. According
to the Policies.doc document, Metaphor credit card
users must pay off their balances every month.
9 Change the filter to:

(NEWBAL > 1000) OR (PASTDUEAMT > 0) OR
(FINCHG > 0)
Filters used in this session can be recalled from the

drop-down list in the Filter field.
38
ACL in Practice
Exercises
Exercises
1 Extract the records with a NEWBAL value greater
than 1000 to a new table for later analysis.
Chapter review
In this chapter, you:
Calculated Metaphors total credit card financial

liability.
Found the credit card numbers that are due to

expire soon and extracted them to a text file.
Found transactions linked to a particular

customer.
Checked the NEWBAL field to determine how

many transactions exceeded particular limits.
2 What is the difference between the total number of

transactions and the number of transactions with a
NEWBAL value greater than 2000?
3 Of the credit card numbers that have a NEWBAL

value greater than 2000, how many have a
PASTDUEAMT greater than zero?
4 Check for card numbers with a zero NEWBAL value.

Export these to a text file.
5 Check the Credit_Card_Metaphor table for all card

numbers with a negative NEWBAL value. Export
these to an Access file.
6 Under the terms of the credit card program, Metaphor

can avoid finance charges if employees submit their
receipts on time to the accounting department.
Metaphor wants to determine how much money it
loses unnecessarily from finance charges. Total the
FINCHG field. Also, how many credit card holders
have a FINCHG value that is greater than $10.00?
7 Metaphor wants to know the total of all current

balances. In the Credit_Card_Metaphor table, total
the NEWBAL field.
8 Metaphor is considering reducing its upper credit

limit. How many credit card numbers have credit
limits exceeding $10,000?
ACL in Practice
39
40
ACL in Practice
C HAPTER 5
E XAMINE EXPENSE PATTERNS
Determine information about expenses
In this chapter
Find the total expenses for each
category code
Determine specific transaction
information
In the previous chapters, you worked with ACL to gain familiarity with your
data and you performed some preliminary analysis of the data. Now the
accounting department at Metaphor Corporation wants you to provide
information about spending patterns in the April credit card transactions. If
you can develop a useful analysis plan, you will be asked to perform analyses
for other months.
Once you provide information about the spending patterns, Metaphor might
be able to economize by choosing a single vendor in some cases, or by
obtaining bulk discounts, or by further limiting the acceptable categories of
transactions. You will also look at expenses on a per-customer basis so that
Metaphor can use the results to determine whether the money is being spent
effectively.
41
Find the total expenses for each

category code
You begin your analysis by summarizing the total amount
of all the transactions for each category code. Your plan is
to look at these results to get ideas for further analysis.
Tasks
Summarize the total expenses for each category code
Isolate transactions for airfare, hotel stays, and rental cars
Isolate particular recreational expenses
Summarize the total expenses for each category

By summarizing the transaction amounts for each
merchant category code, you can see how much money
was spent on specific airlines and specific hotel chains. All
restaurant expenses, on the other hand, will be
categorized under a single code. Some Merchant
Category Codes are specific, while others are broad.
Summarize transaction amounts by category code

2 Select Analyze Summarize. The Summarize dialog
box appears.
3 In the Summarize On list, select CODES.

4 In the Subtotal Fields list, select AMOUNT. You want
to total the amount spent on each category code.
5 In the Other Fields list, select DESCRIPTION. You

want to see what the codes mean in the new table.
6 On the Output tab, select File. In the Name text box,

enter Trans_by_Code and click OK.
The summarized results are extracted to the
Trans_by_Code table.
42
ACL in Practice
Find the total expenses for each category code
View the new table. Note that the COUNT field indicates
the number of transactions per code. You can do a Quick
Sort Descending on the AMOUNT field to see where the
most money was spent. You can also use this table for
further analysis.
Isolate transactions by category

Metaphor wants you to look closely at the total amount
spent on out-of-town stays in April, specifically on
airfare, accommodation, and rental cars.
Hotels, motels, rental cars, and airfare fall exclusively in
the code range 3000-3750. All transactions with these
codes should relate to travel by Metaphor employees.
You can isolate these expenses by filtering and displaying
all transactions with codes between and including 3000 to
3750.
One way to isolate this information is by using the
BETWEEN( ) function on the CODES field.
BETWEEN( ) works on numeric, date, and character
data, and it takes the following form:
BETWEEN(value,min,max)
This summary table is useful for showing the amounts

spent on airlines, hotel chains, rental car agencies, and
other categories. However, it does not show you the total
amount spent on each of airfare, hotels, and rental cars,
nor does it tell you the total amount spent on travel.
Often, when you first perform an analytical task, you do
not get precisely the results you want. However, you can
use those results to help you find a way to achieve more
focused results.
You will use the BETWEEN( ) function to capture all

transactions with codes in the range 3000 to 3750. In the
Trans_April table, the CODES field has extra spaces, so
you have to use the ALLTRIM( ) function to trim the
CODES field.
Use the BETWEEN( ) function to find transactions in a
given range

2 Click the Edit View Filter button. The Edit View
Filter dialog box appears.
3 Enter the following filter in the Expression box:

BETWEEN(ALLTRIM(CODES),"3000","3750")
4 Click Verify. A message box indicates that the

expression is valid. Click OK.
ACL in Practice
43
If the expression is not valid, look at it carefully again,

particularly the quotation marks, and make the
necessary changes.
5 Click OK. The Edit View Filter dialog box closes. You
now have a filtered view of the transactions that
displays transactions with Merchant Category Codes
between 3000 and 3750.
You can now perform further analyses of these filtered

transactions. For example, you can determine the total
spent on airfare alone. Because the Merchant Category
Codes for airfare are between 3000 and 3299, you can use
the BETWEEN( ) function to find values between 3000
and 3299.
6 Select Analyze Count Records. Click OK. How

many records fit this criteria?
7 Select Analyze Total Fields. The Total dialog box

appears. Select the AMOUNT field. Click OK.
ACL displays the total of all transactions for airfare,
accommodation, and rental cars. You should have a
total of 47,248.27.
To facilitate further analysis of these transactions, you
can extract the filtered transactions to a new table.
8 Select Data Extract Data. The Extract dialog box

appears.
9 Ensure Record is selected. Enter

Trans_April_Travel in the To field. Click OK.
44
ACL in Practice
Find the total expenses for each category code
Isolate particular recreational expenses

So far you have looked at travel-related expenses for April.
Metaphor now wants you to look at the total expenses
related to golf and drinking establishments. You need to
isolate those transactions with a Merchant Category Code
that is related to golf or drinking establishments.
5 Count the records. The status bar indicates that

11/281 records fit the MATCH( ) parameters.
The MATCH( ) function provides a way to apply multiple

filters to a table. The filters must all relate to one field,
which is CODES in this example. You could achieve the
same results by using a filter with multiple OR statements,
but the filter would be more difficult to set up and
understand at a glance.
Again, you use ALLTRIM( ) to trim the codes before
comparison.
Use the MATCH( ) function to isolate recreational expenses
1 Open Trans_April.
You can use the Total command to find the total spending
on these two categories. You could apply further filtering.
You could also extract the transactions to a new table.
Filter dialog box appears.
3 In the Expression box, enter the following expression:

MATCH(ALLTRIM(CODES),"5813","7992")
Note that the syntax for the MATCH( ) function is:

MATCH (comparison_value, test1, test2
<,test3 ...> )
5813 is the code for Drinking Establishments.

7992 is the code for Public Golf Courses.
You could add more MATCH( ) parameters to search
for more codes if you wanted.
4 Click OK.
ACL in Practice
45
Exercises
1 Total the expenses for airfare, hotel stays, and rental
cars separately.
2 Total the expenses for all British Airway flights.

3 Total food-related expenses. How much money did
the company spend on restaurants, fast-food
restaurants, and catering, both separately and
together?
4 Use MATCH( ) to find transactions billed to internal

departments. Internal departments have customer
numbers that begin with zero.
5 Determine what percentage of the total transaction

amounts is billed to internal departments, and what
percentage is billed to external clients.
List the amounts for each card

number
Now that you have a complete list of transactions, you
want to determine the total transaction amounts for each
credit card number for the month of April. You can get
this information by using the Classify command.
Use the Classify command
1 Open the Trans_April table. Remove any filters, if

applicable.
2 Select Analyze Classify.

3 Select CARDNUM from the Classify On list. Click
AMOUNT in the Subtotal Fields list.
4 On the Output tab, select File. Enter

Trans_April_by_CCN in the Name field. Click OK.
View the new table.

The Trans_April_by_CCN table lists the
accumulated transaction amounts per card number.
The COUNT field lists the number of transactions per
card number.
46
ACL in Practice
List the amounts for each card number
5 Right-click the Percent of Count column and select

Remove Selected Columns. Do the same for the
Percent of Field column.
Exercise
Use the Classify command to create a table that lists
the total transactions on a per-customer-number
basis.
to save.
ACL in Practice
47
Chapter review
48
Determined the total transaction amounts for

each merchant category code.
Determined transaction amounts for specific

expense categories.
Determined the total transaction amounts per

credit card number.
ACL in Practice
C HAPTER 6
A NALYZE TRANSACTIONS
Identify unacceptable transactions
In this chapter
Determine which transactions
are unacceptable
Identify the employees who
made unacceptable transactions
Report your findings
So far you have performed a preliminary investigation and analysis of the

Metaphor data in your ACL projects. You now determine whether Metaphor
employees are using their company credit cards to make unapproved
purchases, and whether they are using their company credit cards to buy
specifically forbidden items and services.
49
Determine which transactions

are unacceptable
Metaphor Corporation has created three categories of
transaction codes, based on the Merchant Category
Codes list. These include:
Acceptable
Unacceptable
Conditional (requires prior approval)
About the Join and Relations commands

When your project includes multiple data tables, you
often want to find whether particular field values exist in
more than one table. Examples:
If you manage a travel agency, you might have a list

of customers who purchased airfare and a list of
customers who purchased hotel rooms. You can
determine how many customers purchased both
airfare and hotel rooms from your agency.
If you work in marketing at a department store,

you might have a list of customers who purchased
over $2000 last year at your store and a list of
customers who obtained refunds of over $1000.
You can determine which customers purchased
over $2000 last year and also obtained a refund of
over $1000.
One table lists the acceptable codes and one table lists the
unacceptable codes. Codes that are not listed in either
table are conditional.
To find all unacceptable transactions, proceed through
the following tasks.
Tasks
Check the code tables for obvious errors.
Identify transactions that have unacceptable codes.
Identify the employees who made the unacceptable transactions.
You can use the Join and Relations commands to compare

tables and to combine data from tables. With these
commands, you can:
Find field values that exist in one table, but not in

another.
Find field values that exist in both tables.
Combine specified fields from two source tables.
Create a report of your results.
The Join command creates a new table that contains data

from both of the original tables. The resulting table looks
like any other data table in your project.
The Relations command does not physically combine the
tables, but creates a virtual join among two or more tables.
You can use the fields from the related tables as if all of the
fields belonged to the same table. You can create
expressions that use fields from any of the related tables.
You can add fields from one related table to the view of
50
ACL in Practice
Determine which transactions are unacceptable
another related table so that you can see fields from both
tables in a single view. If you want, you can then extract
any or all of the fields from the related tables to a new
table.
Because the Join command creates a new table based on
the two original tables, you generally use the Relations
command instead of Join if the original tables are very
large.
To learn more about Join and Relations, select Help Index and
look up combining tables.
Check the code tables for obvious errors

Before you can use the code tables to check the
transactions, check the tables for obvious errors. You will
use the Join command to determine whether the
Acceptable_Codes table and the Unacceptable_Codes
table have any codes in common.
Before you use the Join command, you must ensure that
the key fields of both tables are character fields that are the
same length.
Compare the length and type of the key fields
1 Open the Unacceptable_Codes table.

appears.
The list of fields shows that the data type of the Codes
field is ASCII. The field length is equal to either the
length of the field name, or the length of the longest
value in the field, whichever is longer.

4 Open the Acceptable_Codes table.
appears.
The list of fields shows that the data type of the Codes
field is ASCII, but the field length is 50.
The Merchant Category Codes are each four characters
long. Each of these Codes fields is padded with extra
spaces. You must harmonize the two fields by creating
computed code fields in each table. These new fields will:
ACL in Practice
Contain the four digits of the Codes field as

character data.
Have a field length of 4.
51
You can then perform the join successfully by using the

computed fields as the key fields.
Create a computed field in each of the Acceptable_Codes
and Unacceptable_Codes tables
1 Open the Acceptable_Codes table, if not already

open.
2 Open the Table Layout window, and click the Add a

New Expression
button.
3 In the Name text box, enter
to save.
9 Open the Unacceptable_Codes table. Repeat steps 2

7, but call the computed field
Codes_Sub_Unacceptable.
Now that the key fields are harmonized, you can join the
tables by using the Codes_Sub_Unacceptable field from
the Unacceptable_Codes table and the
Codes_Sub_Acceptable field from the
Acceptable_Codes table.
Codes_Sub_Acceptable.
4 Click f(x) to open the Expression Builder. In the

Expression text box, enter:
SUBSTR(ALLTRIM(Codes),1,4)
The ALLTRIM( ) function trims any extraneous

spaces that occur on the left or the right of the
Codes field.
The SUBSTR( ) function reads the first four
characters of the Codes field.
5 Click the green checkmark. The new field,

Codes_Sub_Acceptable, appears in the Table Layout
window. Verify that the category is C for character and
that the field length is 4.

You can now add the Codes_Sub_Acceptable field to
the view.
7 Right-click the Codes column header. Click Add
Determine whether the tables have any codes in common
1 Open the Unacceptable_Codes table.

2 Select Data Join Tables. The Join dialog box
appears.
3 In the Secondary Table list, select

Acceptable_Codes.
4 Select Codes_Sub_Unacceptable as the Primary

Key. Select Codes_Sub_Acceptable as the Secondary
Key.
5 Select Codes_Sub_Unacceptable and Description in

the Primary Fields list. If there are any codes that
exist in both tables, you will be able to see their
description.
6 Select Presort Secondary Table.

7 Enter Codes_in_both_tables in the To field.
This is the name of the new table that will be created.
Columns. The Add Columns dialog box appears.

Double-click Codes_Sub_Acceptable and click OK.
The Codes_Sub_Acceptable column appears in the
view.
52
ACL in Practice
Identify unacceptable transactions

The Trans_April table contains these fields:
8 On the More tab, select Matched Primary Records,

and ensure that no other Join Categories options are
selected. Click OK.
9 ACL displays the new table. It should be empty.

Although you can see that the table is empty, you
should still count it in order to record the results in the
command log and leave an audit trail.

Because there are no records in the
Codes_in_both_tables table, you know that no codes
exist in both the Unacceptable_Codes and
Acceptable_Codes tables. The original code tables have
no obvious errors, so you can use them for your analysis.
CARDNUM
AMOUNT
TRANS_DATE
CODES
CUSTNO
DESCRIPTION
The Unacceptable_Codes table contains the following

fields:
Codes
Description
Codes_Sub_Unacceptable (a computed field)
To identify unacceptable transactions, you need to relate

the Trans_April table with the Unacceptable_Codes
table. However, you must first ensure that the key fields for
each table both have a data type of ASCII and that they
both have the same lengths.
In the previous section, you created a computed field
called Codes_Sub_Unacceptable in the
Unacceptable_Codes table, and you know that the length
of the Codes_Sub_Unacceptable field is 4 and that the
category is C for character. You need to check the CODES
field in the Trans_April table to check its data type and
length.
View the table layout for Trans_April

2 Select Edit Table Layout.
ACL in Practice
53
The list of fields shows that the CODES field has a

category of C, but it has a field length of 7.
You must create a computed field in the Trans_April table
that trims the CODES field and has a Length of 4. You can
then use this new computed field as the key field for
relating Unacceptable_Codes and Trans_April.
7 Right-click the CODES column header. Click Add

Double-click Codes_Sub_April. Click OK.
Create a computed field in Trans_April
1 In the Table Layout window, click the Add a New

Expression
button.
2 Enter Codes_Sub_April as the name.

3 Click f(x). The Expression Builder appears.
4 In the Expression box, enter:
SUBSTR(ALLTRIM(CODES),1,4)
5 Click OK.
Next, determine whether any of the April transactions

have unacceptable codes. Your first step is to isolate
unacceptable transactions by comparing the transaction
codes in both tables.
Use the Relations command to find unacceptable
transactions
1 With the Trans_April table active, select Data

Relate Tables. The Relations dialog box appears.
6 Click the green checkmark. Close the Table Layout
If the Getting Started message also appears, select Do

not show this dialog again and click OK.
window.
2 Click Add Table.
54
ACL in Practice
3 Double-click Unacceptable_Codes. The

Unacceptable_Codes table appears in the Relations
dialog box.
8 Expand the Name column so that you can see the field
names. Double-click
Unacceptable_Codes.Codes_Sub_Unacceptable.
To arrange the tables within the Relations dialog box, drag

the tables with your mouse or right-click and select Arrange
Tables.
4 Click and drag Codes_Sub_April from the

Trans_April table to Codes_Sub_Unacceptable in
the Unacceptable_Codes table.
9 Click OK.
Now that you have the Codes_Sub_Unacceptable field
from the Unacceptable_Codes table in the view, the
records without unacceptable codes display blanks in the
Codes_Sub_Unacceptable field.
If you link the wrong fields, right-click the line and
select Edit Relation. Select the fields that you want to
link, and click OK.
5 Click Finish.
Next, you will add the Codes_Sub_Unacceptable
field from the Unacceptable_Codes table to the view.
6 Right-click the Codes_Sub_April column header.

Click Add Columns. The Add Columns dialog box
appears.
7 In the From Table drop-down list, select

Unacceptable Codes. The fields in the
Unacceptable_Codes table appear in the Available
Fields list.
You can apply a filter to display unacceptable transactions

only.
Apply a filter to display unacceptable transactions
1 Select and right-click any blank entry in the

Codes_Sub_Unacceptable field. Select Quick Filter
Not Equal.
You have filtered out all of the records that have blank
entries in the Codes_Sub_Unacceptable field. Now
ACL displays records that have unacceptable codes
only.
2 Remove the Codes_Sub_April and the CODES fields

from the view.
3 Select Analyze Count Records. Click OK. There are

7/281 unacceptable transactions.
Next, you extract the records to a new table.
ACL in Practice
55
Extract the records to a new table
1 Select Data Extract Data. The Extract dialog box

appears.
2 In the Extract dialog box, select Record. In the To

field, enter Trans_April_Unacceptable. Click
OK. Click Yes when you are asked if you want to save
the changes to Trans_April.
The new table appears in the view. Adjust the column
widths as necessary and rearrange the columns as
desired.
Use the Count command to count how many
unacceptable transactions there are, in order to write
this number to the log and create an audit trail. You
should have seven unacceptable transactions.
By looking at the results, you can see that four
unacceptable transactions took place on April 14.
You can create a date filter and count the number of
records to write this information to the log.
to save.
Identify the employees

Now that you have confirmed that some transactions do
have unacceptable codes, you determine which employees
made these purchases. You can also identify the
department to which these employees belong.
You can obtain information from several tables and
combine it in a single view. The goal is to report:
Employee name
Credit card number
Department
Transaction amount
Transaction date
Transaction code
Transaction code description
Tasks
Determine the department of employees whose cards were used for
unacceptable transactions.
Create a list of employees whose cards were used for unacceptable
transactions.
Create a report of the employees whose cards were used for
unacceptable transactions.
56
ACL in Practice
Identify the employees departments

Your first step is to associate each employee with his or her
department.
5 Click f(x). The Expression Builder appears. Enter the

following expression:
SUBSTR(ALLTRIM(Dept_Code),1,1)
The Company_Dept table contains department

codes and names, such as 800000 for Consulting.
6 Click OK. Click the green checkmark.

The Employees table contains the EmpNo field,

which begins with the first digit of the department
code.
You now have a new computed field called Dept_Code_1,

which you can now add to the view.
By comparing the first digit of the employee code to the

first digit of the company department code, you can
identify the department to which the employee belongs.
To make this comparison, you create a computed field in
each table that takes the first digit of the Dept_Code
values. You then compare this digit to the first digit of the
EmpNo in the Employees table.
You start by checking the data types of these fields on the
Edit Fields/Expressions tab of the Table Layout window.
Create a computed field that contains the first digit of the
Dept_Code field
Add the new computed field to the view
1 Right-click the Dept_Code column header. Select

Add Columns.
2 Double-click Dept_Code_1. Click OK.

The digit in the Dept_Code_1 column matches the
first digit of the Dept_Code column.
You can remove the Dept_Code column because the
new column has the relevant information.
3 Right-click the Dept_Code column and select

Remove Selected Columns. Click Remove.
1 Open the Company_Dept table.

appears. On the Edit Fields/Expressions tab, ensure
that the Dept_Code field has a data type of ASCII.
Change the type if it is incorrect.
3 On the Edit Fields/Expressions tab, click the Add a

New Expression
button.
4 Enter the name Dept_Code_1.
4 Close the view and click Yes when prompted to save.

Next, you create a computed field in the Employees table
that uses the first digit of the employee number field,
EmpNo.
ACL in Practice
57
Create a computed field that displays the first digit of the

EmpNo field
1 Open the Employees table.

appears.
3 On the Edit Fields/Expressions tab, click the Add a

New Expression
button.
4 Enter the name EmpNo_1.

5 Click f(x). The Expression Builder appears. In the
Expression box, enter:
SUBSTR(ALLTRIM(EmpNo),1,1)
6 Click OK. Click the green checkmark.

Now add this field to the view.
8 Right-click the EmpNo column header. Select Add
5 Click Finish.
Your next step is to add the Dept_Name column to the
view.
6 Right-click the CardNum column header and select

Add Columns. The Add Columns dialog box
appears.

Company_Dept. Double-click the
Company_Dept.Dept_Name field. Click OK.

Double-click EmpNo_1. Click OK.
The new EmpNo_1 column now appears in the view.
Now relate the Employees table to the Company_Dept
table, using the EmpNo_1 field and the Dept_Code_1
field.
Relate the Employees and Company_Dept tables
1 With the Employees table active, select Data Relate

Tables. The Relations dialog box appears.
2 Click Add Table. The Add Table dialog box appears.

3 Double-click Company_Dept.
4 Drag EmpNo_1 in the Employees table to the
The Dept_Name column appears in the view. You are

close to your goal of creating a report. Your next task will
be to extract this view to a new table.
Dept_Code_1 field in the Company_Dept table.
58
ACL in Practice

Extract fields to a new table
1 With the Employees table active, select Data

Extract Data. The Extract dialog box appears.
2 Select Fields. Click the Extract Fields button.

The Selected Fields dialog box appears.
3 Add the following fields to the Selected Fields list:
First_Name
Last_Name
CardNum
Create a list of employees

Your next step is to relate the Employees_Dept table to
the Trans_April_Unacceptable table using the card
number field in each table.
Before you can use the Relations command, you must
harmonize the two card number fields so that you can use
them as key fields. Check the card number fields in the
Table Layout window for both tables.
Compare and harmonize the card number fields
1 With the Employees_Dept table active, select Edit
Company_Dept.
Table Layout.
The fields in the Company_Dept table appear.
The list of fields indicates that the CardNum field has

a field length of 16 and a category of C for character.
5 Add Company_Dept.Dept_Name to the Selected

Fields list.

3 Open the Trans_April_Unacceptable table.
4 Select Edit Table Layout.
The CardNum field has a field length of 19, which
means that this field contains extra spaces. Therefore,
you must trim this card number field.
6 Click OK.
7 In the To field, enter Employees_Dept. Click OK.
Click Yes when ACL prompts you to save.
The new Employees_Dept table opens in the view.
ACL in Practice
5 Click the Add a New Expression button.

6 Enter CardNumTrim as the Name.
7 In the f(x) box, enter this expression:
SUBSTR(ALLTRIM(CardNum),1,16)
8 Click the green checkmark.

59

Create a detailed view of the unacceptable transactions
1 With the Trans_April_Unacceptable table active,

select Data Relate Tables.
Trans_April_Unacceptable already has a relation
with Unacceptable_Codes.
2 Select Add Table. The Add Table dialog box appears.

Double-click Employees_Dept.
3 Drag CardNum from the Employees_Dept table to

CardNumTrim in the Trans_April_Unacceptable
table.
Employees_Dept.Dept_Name
Click OK.
7 Arrange the fields in the view so that they appear in

the following order:
First_Name
Last_Name
Dept_Name
AMOUNT
DESCRIPTION
TRANS_DATE
CARDNUM
Codes_Sub_Unacceptable
CUSTNO
You can now see the list of employees whose credit cards
were used to charge unacceptable transactions. You can
rearrange and adjust the fields as necessary so that the
view is attractive.
4 Click Finish.
Three tables, Trans_April_Unacceptable,
Unacceptable_Codes, and Employees_Dept, are
now related.
In the view, you now add all of the columns that you
want to include in the final report.
5 Right-click the CardNum column header. Select Add


Employees_Dept. Add these fields to the Selected
Fields list:
60
Employees_Dept.First_Name
Employees_Dept.Last_Name
Of course, you cannot assume immediately that the

employees listed are responsible for the unacceptable
transactions because it is possible someone else used their
credit cards. Nevertheless, you can create a report of the
results, and then the staff at Metaphor Corporation can
investigate the matter further.
ACL in Practice
Report your findings

ACL provides two ways to create a report. You can
generate a standard ACL report, or you can create a report
that you can view in Crystal Reports. In this procedure,
you use ACLs standard report capability.
6 In the Name text box, enter

Unacceptable_Trans_Apr_2003. Click OK.
7 In Windows Explorer, locate the folder C:\ACL

Data\Sample Data Files and double-click
Unacceptable_Trans_Apr_2003.html. The report
opens in your Web browser.
To learn more about using Crystal Reports to generate a report,

select Help Index and look up Crystal Reports.
Create a standard ACL report
1 Select Data Report.

The Report dialog box appears.
2 Enter a descriptive Header, such as:

Unacceptable Transactions at Metaphor
Corporation in April 2003

The report appears on the Do tab.
Next, you save the report as an HTML file.
4 Select Data Report.

5 On the Output tab, select File. Select HTML Text File
from the File Type drop-down list.
ACL in Practice
61
Chapter review
Exercises
1 Create a report that lists all acceptable transactions.
The procedures will be the same as for the report that
lists unacceptable transactions, except you will use the
Acceptable_Codes list as the basis for comparison.
Using Relations, create a view that displays the
following information:
Employee name
Credit card number
Department
Acceptable transaction amount
Transaction date
Transaction code
Transaction code description
Examined the tables that contain acceptable and

unacceptable Merchant Category Codes to ensure
that no codes are listed in both tables.
Created a table with unacceptable transactions

only.
Linked the employees who made unacceptable

transactions to the employees departments.
Created a report that provides a detailed

description of all unacceptable transactions in
April.
Extract this view to a new table

Acceptable_Trans_by_Emp_April, and export the
new table to an Excel file.
2 Remember that some transactions have neither

acceptable nor unacceptable codes. These are the
transactions that require approval, but are not
forbidden. Create a table of transactions that are
neither acceptable nor unacceptable. How many
transactions are in this category?
3 Extract all of the transactions that exist in the

approval-required category to another table.
4 Refer to the Trans_April,

Trans_April_Unacceptable, and Employees tables.
What percentage of employees with credit cards made
unacceptable transactions?
5 How would you use ACL to e-mail the results of your

report?
62
ACL in Practice
C HAPTER 7
V ALIDATE , CORRECT , AND EXTRACT DATA
Repair your data so it can be analyzed
In this chapter
Organize your project with
folders
Expand your project to include
May transactions
Combine May transactions into a
single table
Correct card number and
employee lists
So far, you have planned a project, created a project, added data to a project,
and analyzed it. You based your analysis on Metaphors April 2003 credit card
transactions.
Now you have been asked to analyze the May 2003 transactions.
Unfortunately, these transactions are in two separate files, and the data is not
as neat as the April 2003 transactions. You must perform additional steps to
get the data into a state that you can analyze readily. Once you repair the
transaction data, you extract the data to new tables.
63
Import the May transaction data

You already have a project for Metaphor analysis in ACL.
You created a project that contains all of the April
transactions. Now you plan to add May transactions too.
Tasks
Create folders for April and May transactions.
Import worksheets from the Trans_May.xls Excel workbook.
Compare layouts of the two transaction tables.
Use folders to organize your project

To keep your project organized, you will move your April
work into another folder. You can then set up another
folder for your May work. All of the work can then be
viewed as needed, but the Overview will not be cluttered
with items that you are not using.
Create a folder called April 2003
1 Select File New Folder. A folder appears. Enter the

name April 2003 for the folder. Press ENTER.
2 Select and drag all of the tables to the April 2003

folder, as you would in Windows Explorer.
3 Click ACL_Tutorial_Metaphor.ACL. Select File

New Folder. Name the folder May 2003 and press
ENTER.
64
ACL in Practice
Import the May transaction data
Import transaction data for May 2003

The Trans_May.xls workbook contains two worksheets
with your transaction data for May 2003. Refer to
Chapter 3: Set up your project if you are uncertain
about how to import data.

4 Select Data Verify, then verify all of the fields. You
should have zero validity errors.
5 Repeat the same procedures for the Trans2_May

table.
Import the Trans_May.xls Excel workbook
1 Right-click the May 2003 folder. Select New Table.

The Data Definition Wizard appears. Continue
through the wizard, and select Trans_May.xls.
2 Because ACL saves the data from each worksheet as a

separate table, you need to import Trans_May.xls
once for each worksheet. On the Data Source screen
select the first worksheet and enter a value in the
Maximum Character Field Length box.
3 Save the table as Trans1_May. Return to step 1 and

save the second worksheet as Trans2_May.
Edit the table layout for both new tables
1 With the Trans1_May table active, select Edit Table

Layout. The Table Layout window appears.
2 Edit the AMOUNT, CARDNUM, and CODES fields:
Double-click the AMOUNT field. Select

NUMERIC from the Type drop-down list. Enter 2
in the Dec field so that AMOUNT has two decimal
places. Select -999999.99 from the Format
drop-down list. Click the green checkmark.
Double-click the CODES field, change the data

type to ASCII, and click the green checkmark.
Double-click the CARDNUM field, change the

data type to ASCII, and click the green
checkmark.
ACL in Practice
65
Compare and correct the table layouts

Before you can combine your tables, the Start, Type, and
Length values for each field in each table must be exactly
the same. To compare the tables, you can use DISPLAY on
the command line.
5 In the Command Line text box, enter DISPLAY.

Press ENTER. ACL displays the table layout.
Most activities in ACL can also be accomplished by typing

ACLScript commands on the command line. ACLScript is the
language for writing scripts and executing commands in ACL.
6 Compare the contents of both Display tabs. The table

Compare the layout of the transaction tables
1 Open the Trans1_May table. Select Window Show

Command Line. The command line appears.
2 In the Command Line text box, enter DISPLAY.

Press ENTER. ACL displays the table layout.
layouts are identical except for one inconsistency: in

Trans1_May, AMOUNT has a field length of 16, but
in Trans2_May, AMOUNT has a field length of 15.
You can resolve this problem by correcting the layout
of Trans1_May.
Correct the layout of Trans1_May
1 Open Trans1_May.
2 In the view, right-click the AMOUNT column and
select Quick Sort Descending.
3 On the Display tab, click the push-pin
button to
ensure that ACL displays subsequent command results
in addition to the pinned results.
4 Open the Trans2_May table.
The view shows that the highest AMOUNT value you

have in Trans1_May is 1983.04. This value does not
require 16 characters. You can safely change the
AMOUNT field length to 15 so that it matches the
AMOUNT field in the Trans2_May table. AMOUNT
starts at position 96, and no field comes after
AMOUNT, so you will not affect the positions of the
other fields by changing AMOUNT.

appears. Double-click AMOUNT. Change Len to 15.
Click the green checkmark.
66
ACL in Practice
Combine and verify the tables

In this section, you combine the transactions from both
May tables into one table.
You can combine information from two or more tables by
using the following methods:
Join The Join command creates a new table that

includes selected fields from both files.
Relations The Relations command does not
create a new table, but the fields in each of the
related tables can be analyzed as if they were in the
same table.
Extract-and-append You can extract records

and append them to an existing table.
Merge You can merge fields from two tables into

one table.
You have two credit card transaction tables, Trans1_May

and Trans2_May, with exactly the same layout. To
analyze them as one unit, you combine them.
Tasks
Use extract-and-append to combine the May transaction tables.
Verify the combined table.
Check the table for duplicate transactions.
Combine the May transaction tables

Using the extract-and-append method, you can extract
Trans1_May into a new table.
You can then extract the information in Trans2_May and
add it to the end of the new table. The new table will
contain the combined information from both tables.
You could use other methods to combine the data, but the
extract-and-append method is ideal for tables with
identical layouts.
Use the extract-and-append method to combine two tables
1 With the Trans1_May table active, select Data

Extract Data.
2 Select Fields. Click Extract Fields. The Selected

Fields dialog box appears.
3 In the Available Fields list, click the Start column

header so that all of the fields are listed in sequential
order starting with CARDNUM.
4 Click Add All. Click OK. You are extracting all of the
fields in the order listed.
5 Enter Trans_May_All in the To field. Click OK.

ACL creates a new table, Trans_May_All, which
contains the data from Trans1_May.
Now append the Trans2_May data to the
Trans_May_All table.
6 Open Trans2_May. Select Data Extract Data.

7 Select Fields. Click Extract Fields. The Selected
Fields dialog box appears.
8 In the Available Fields list, click the Start column

header so that all of the fields are listed in sequential
ACL in Practice
67
order starting with CARDNUM. This order is exactly

the same as the order that you used for Trans1_May.
9 Click Add All. Click OK.

10 Enter Trans_May_All in the To field.
11 On the More tab, select Append To Existing File.
12 Click OK.
ACL opens Trans_May_All. You can now work with this
table in ACL as you would any other table.
Verify the new table

Next, you use the Verify and Count commands to ensure
that the two tables have been appended correctly.
Verify the new table
1 Select Data Verify.

2 Click Verify Fields. Click Add All. Click OK.
3 On the Output tab, ensure that Screen is selected.
Click OK.
You should get the result: 0 data validity errors detected.
Compare the Count in each table
1 In the Trans_May_All table, select Analyze Count

Records. Click OK. You should have 200 records.
2 Count the records in the Trans1_May table.

You should have 86.
3 Count the records in the Trans2_May table. You

should have 114.
The extract-and-append procedure appears to have been
successful.
68
ACL in Practice
Check the combined table for duplicates
Exercises
Always check for data-entry errors when you combine

two tables. Duplicate entries can indicate that someone
entered the same transaction twice in the original Excel
worksheets.
1 Run the Profile command on the Trans_May_All
Duplicate transactions can also indicate fraud. Employees

can split credit card transactions to avoid scrutiny.
Instead of a $2000 purchase, the employee might have two
separate transactions of $1000, for example. Alternatively,
merchants might charge twice for the same product or
service.
3 Export a table to XML that lists the employee name,
table. Note the highest and lowest transaction values.
2 Check for duplicates in the CUSTNO field. Classify

the table on the same field. Are the results significant?
card number, and total dollar value of associated
transactions per-card in May. View the file in Internet
Explorer.
Use the Duplicates command to look for transactions of

identical amounts.
Check the Trans_May_All table for duplicates
1 Open the Trans_May_All table.

2 Select Analyze Look for Duplicates. The
Duplicates dialog box appears.
3 Select AMOUNT in the Duplicates On list.

4 Click List Fields. Click Add All. Click OK.
5 On the Output tab, ensure that Screen is selected.
Click OK.
6 View the results. You are checking for duplicate

AMOUNT values.
You should have a transaction for a computer software
store and one for Southwest in the same amount. You
should also have two transactions at the Hilton one day
apart in the same amount.
What do you think is the possible significance of these
duplicate amounts?
ACL in Practice
69
Isolate and correct validity

errors
In the last section, you verified the data that you
imported. However, verifying the data is different from
validating it. For example, you determined that the
CARDNUM field consisted only of characters, so that the
data is ASCII text. However, you still must determine
whether the card numbers are valid, which means that
each entry consists of 16 digits and no other characters.
In this section, you create:
A filter to display only the valid entries.
A filter that displays only the invalid entries.
If some entries are invalid, you can either correct the

problems, or filter out the records that contain those
entries.
Tasks
Determine whether all the card numbers have exactly 16 digits.
Make a 16-digit credit card field.
Extract repaired card number records to a new file.
Correct the new list of employees.
Check the format of card numbers

You need to determine which card numbers in the table
are in an invalid format. To do this, you apply a filter to the
CARDNUM field that uses a nested function that consists
of ALLTRIM( ) and MAP( ):
ALLTRIM( ) removes all trailing blanks from each

card number entry.
MAP( ) evaluates whether each card number,

trimmed of trailing blanks, has exactly 16 digits.
Apply a filter to show invalid card numbers
1 With the Trans_May_All table active, click the Edit

View Filter
button.
The Edit View Filter dialog box appears.
2 In the Expression box, enter the following expression

all on one line:
MAP(ALLTRIM(CARDNUM),
"9999999999999999")
The string of sixteen nines indicates that the filter

determines whether each CARDNUM value consists
of sixteen digits.
3 Click OK. ACL applies the filter.

4 Count the number of records. As you can see, 195/200
records remain in the view, which means that 5
records do not meet the criteria.
5 In the Filter text box, add NOT to the beginning of

the expression:
NOT MAP(ALLTRIM(CARDNUM),
"9999999999999999")
Press ENTER. NOT causes ACL to display the invalid

entries.
70
ACL in Practice
Isolate and correct validity errors
6 Count the records. You should have 5/200 visible.

The invalid card numbers contain either extraneous
dashes or spaces. In this case, you can see all of the invalid
entries at a glance, but in a large project, you might not be
able to do so.
Create a 16-digit card number field

Now you compare these card numbers to another list of
card numbers that do not contain dashes and spaces. To
do this, create a computed field in Trans_May_All in
which all of the card numbers are sixteen digits long with
no spaces or dashes.
The field that you create must also contain all of the valid
card numbers without changing them. Invalid card
numbers must be corrected without altering the digits.
Create a computed field with corrected card numbers
1 In the Trans_May_All table, click the Remove Filter

button to remove the filter. The view now displays
all of the records.

appears.
2 Click the Add a New Expression button.

3 Enter CCN_Corrected in the Name field.
4 Click the f(x) button. The Expression Builder
appears. The expression that you create will populate
the CCN_Corrected field.
5 Enter the following expression:

ALLTRIM(INCLUDE(CARDNUM,"0123456789"))
6 Click OK. The Edit Fields/Expressions tab appears

again.
7 Click the green checkmark.

You can now add the new field to the view.
ACL in Practice
71
Add the CCN_Corrected computed field to the view
1 Right-click the CARDNUM header. Select Add

2 Double-click CCN_Corrected. Click OK.

3 Remove the CARDNUM column.
Because you now have the corrected card number
field, you no longer need to see the CARDNUM field.
Now you apply a filter to ensure that the
CCN_Corrected field contains only digits.
4 In the Filter text box, enter:

MAP(CCN_Corrected,"9999999999999999")
Press ENTER.
72
Extract the corrected records to a new table

Now that you have created a field that contains only valid
card numbers, your next step is to extract the records to a
new table. You then use this new table for further analysis.
Extract the corrected records to a new table
1 Select Data Extract Data.

2 Select Fields. Click the Extract Fields button. The
Selected Fields dialog box appears.
3 Click Add All. Because CARDNUM contains invalid

entries, you do not want to include that field in your
new table. Remove CARDNUM from the Selected
Fields list. Click OK.
5 Count the number of records. All 200 are visible.

6 Click the Remove Filter button to remove the filter.
4 In the To text box, enter the file name
Your next step is to extract this information to a new table.
ACL opens Trans_May_All_CCN_Corrected.
Trans_May_All_CCN_Corrected. Click OK.
ACL in Practice
Isolate and correct validity errors
Export the combined transaction table to Excel

You have created and checked a table that combines the
two transaction tables that represent May 2003 credit card
transactions. You might want to have this data in Excel
format as well.
Export the Trans_May_All_CCN_Corrected table to Excel
1 With the Trans_May_All_CCN_Corrected table

active, select Data Export to Other Application.
2 Select Fields.
3 Select the fields you want to export in the order you
want them to appear. The fields will appear in the
Excel worksheet in this order, from left to right.
4 From the Export As drop-down list, select Excel.

5 In the To field enter the filename
Trans_May_All_Corrected. Click OK.
You can also select the path for the file. If you enter just the
filename, the new file will be saved to the last folder that
you used to store your ACL data.
Create an error-free employee record table

For May, you have been given a slightly different list of
employees to use: Employees_May_with_CCN.xls. The
table includes only those employees who have credit cards
with Metaphor.
Import the new table of employees
1 Import Employees_May_with_CCN.xls into the

May 2003 folder and save it as
Employees_May_with_CCN.
2 Scroll through this new table.

You can see that the bottom record has End of File in the
First_Name field, and that all the other field entries for
this record are empty. Clearly, this is not an employee
record, and it should be filtered.
Filter invalid employee records
1 With the Employees_May_with_CCN table active,

right-click the End of File entry.
2 Select Quick Filter Not Equal.

The invalid record is now filtered out.
If you extract the valid data to another table, then you do
not have to worry about preserving the filter on the
original table.
Extract the valid records to a new table
1 Select Data Extract Data.

2 Select Fields button. Click Extract Fields. Click Add
All. Click OK.
3 In the To text box, enter the filename

Employees_May_Corrected. Click OK.
ACL in Practice
73
Ideas for further analysis
Exercises
1 Suppose you wanted all card numbers to have the
form: xxxx-xxxx-xxxx-xxxx. What strategy
would you follow to create a computed field like that?
2 If you had the information available, checking the

Social Security Number (SSN) for each employee
would help determine whether each record was
unique. How would you check for duplicate Social
Security Numbers? How would you ensure that all
Social Security Numbers were in the format of one
long string of digits: xxxxxxxxx?
3 Devise a strategy for analyzing your May transactions

that closely mirrors the strategy you followed for the
April transactions. Find the unacceptable transactions
for this set of transactions.
4 For your own projects, think of various approaches

that improve the quality of your data before you
import it into ACL. Think of ways to test your data
after you import it into ACL.
74
In this tutorial, you have worked through several

examples of how to perform analyses using ACL. What
other analyses could you perform with the data?
Consider these questions. You can likely come up with
numerous ideas that are applicable to your own work.
Do the transactions have unacceptable Merchant

Category Codes?
Do the Merchant Category Codes that are used

match the cardholders responsibilities?
Does the cardholder conduct repeated

transactions at regular intervals with the same
merchant?
Does the cardholder have several transactions

within a brief time period (say 72 hours) with the
same merchant?
Does the cardholder have transactions occurring

on weekends or holidays?
Do an unusual number of transactions have

rounded values, such as $1000, $200, or $450?
Are there credits charged back to a card number,

and what are the explanations for these credits?
Does the cardholder regularly come close to or

exceed his or her credit limit?
Has there been a sudden increase in the

cardholders purchases?
Is the cardholder responsible for charging other

peoples expenses?
Which five or ten employees spent the most? To

which customers did they charge their expenses? Is
ACL in Practice
Chapter review
there any relation between spending a great deal

and charging unacceptable transactions?
ACL in Practice
Determine the total spending by department on a

monthly basis. Use the DeptCode field in the
Employees table.
If an employee has charged air fare, what other
travel-related expenses has he or she charged? For
example, if an employee charges air fare, he or she
is likely to charge a hotel room and transportation.
Is this the case? What are some possible
explanations for the patterns that appear?
If the codes for caterers and bakeries were
unacceptable, how much would money have been
saved?
Chapter review
Imported an Excel workbook with two

worksheets.
Combined information from two tables into one

table.
Corrected a data field that included invalid data.
Extracted the new transactions table to Excel.
Corrected a table by filtering out an invalid record

and extracting the results to a new table.
Created a table that combines all of the CC

transactions into one table.
75
76
ACL in Practice
Index
I NDEX
A
Absolute value
Profile command results 12
Acceptable_Codes.mdb
data file 5
Access. See Microsoft Access data
Account number categories
Metaphor Corporation 4
Allowable expenses
ALLTRIM( ) function
using 43, 45, 52, 54, 57, 59, 70
Analysis strategy
planning 7
Analytical strategy
planning 6
Append option
Extract command 67
Applications
exporting to other 35
Ascending
Quick Sort order 15
ASCII
data type 28
ASCII data
formatting in expressions 36
Audit trail
leaving 53
log tab 15
B
Backup copy
77
ACL project 26
BETWEEN( ) function
using 43
Blank entries
removing 55
Blanks
removing leading and trailing 45, 70
Bonuses
finding total 15
C
Calculated fields. See Computed fields
Categories
data type 29
expense, Metaphor Corporation 4
Cells
blank 55
Character data
applications 28
numerals in 24
Characteristics of data
checking 27
Checkmark
green 17
Classify command
using 46
Code categories
Codes
acceptable 5
common 51
comparing 54, 57
Column titles
changing 25
and field names 25
Column width
changing in view 17
Columns
adding to view 52, 54, 58, 72
changing sort order in view 10, 15
removing from view 55
Combining data
methods 67
Command
defined 15
Command line
using 66
Command results
viewing details 13
Commands
Classify 46
Count 11, 16
DISPLAY 66
Duplicates 13, 69
Export 35, 73
Extract 56, 58
Join 50
Profile 12
Relations 50, 54, 58, 60
running on single fields 11
selecting fields to include 13
Statistics 11
Summarize 42
Total 15, 34
Verify 68
Comma-separated-value files
Index
importing 25
Company_Departments.txt
data file 5
Computed fields
adding to view 57
creating 17, 52, 54
Conditions
multiple 38
Count command
using 11
COUNT field
Summarize command 42
Credit card policies
Credit cards
finding expiry date 34
Credit_Cards_Metaphor.xls
data file 5
Credit_Cards_Metaphor.xls worksheet
importing 23
CSV files
importing 25
Cycle of analysis
typical 6
D
Data
appending 67
combining 50
errors
eliminating 7
exporting 35
filtering 72
format information 7
importing through the wizard 22
integrity testing 13, 27
isolating specified portions 43
modifying with computed fields 28
78
samples for testing 7

sources 6
validity errors, finding 30
verifying type 30
viewing in original format 6
Data Definition Wizard
using 22
Data definitions
modifying 28
Data files
opening 10
tutorial 5
Data format
checking 27
Data type
changing 24, 35
checking 51, 53, 66
verifying 27
Data type categories
Table Layout window 29
Data-entry errors
finding 69
Date data
Statistics command 11
Dates
selecting in Expression Builder 34
Decimals
rounding 17
Default view
displaying 17
Delimited files
importing 23
Descending
Quick Sort order 15
Digits
using as character data 24
DISPLAY command
using 66
Documentation
project 30
Double entries
finding 13
Drill-down filter
command results 13
Duplicate records
finding 13
Duplicates command
using 13, 69
E
Edit Field Properties screen
using 24
Edit view filter
dialog box 16, 17, 34
Employee numbers
finding duplicates 13
Employees.cvs
data file 5
Empty cells
removing 55
End of file record
deleting 73
Entries
duplicate 13, 55, 69
invalid, displaying 70
Errors
data validity 30
data-entry 7, 69
testing for 65
validity 68
Excel. See Microsoft Excel data
Expenses
allowable
isolating 43, 45
summarizing by code 42
Index
Expiry dates
finding 34
Export
using Excel format 73
Export command
using 35, 73
Expression
defined 16
Expression Builder
See also Edit view filter
functions list 35
opening 52
using 17
Expressions
checking syntax 36
creating and saving 17
formatting
character data 36
multiple conditions in 38
Extract command
append option 67
using 44, 56, 58
F
Field length
changing 66
checking 51, 53, 59
Field names
and column titles 25
importing from data file 24
Field order
changing in the view 10
selecting for export 73
Field separators
specifying 24
Field totals
finding 34
Field type. See Data type
79
Field width
adjusting in view 17
Fields
adding to view 52, 54, 58
blank entries in 55
changing order 60
combining from multiple tables 50
computed 28, 52, 54
editing 28
eliminating invalid characters 70
harmonizing 51, 53, 59
key 51, 66
ordering sequentially 67
removing from view 55
sort order 15, 37
verifying content 27
Files
importing 23
Metaphor data 5
Filters
applying multiple 45
creating 16, 34, 72
removing 13, 17
saving 37
setting multiple conditions 38
FIND( ) function
using 35
Folders
creating 64
Format
data
checking 27
data file 6
date expressions 16
expression
verifying 36
invalid 70
Formats
export 35
Functions
ALLTRIM( ) 43, 45, 57, 70
BETWEEN( ) 43
FIND( ) 35
INCLUDE( ) 71
MAP( ) 70
MATCH( ) 45
nesting 70
SUBSTRING( ) 52, 54, 57
using 35
G
Green checkmark
saving changes 17
H
Harmonization
key fields 51, 53, 59
Header rows
using as field names 24
Headings
and column order 10
and sort order 15
History
project 30
HTML
report format 61
Hyperlink
command results 13
I
Identical
entries, finding 69
records, finding 13
Import
CSV files 25
Index
Excel worksheets 23, 65

Microsoft Access data 25
using the wizard 22
INCLUDE( ) function
using 71
Intactness of data
testing 13
Integrity of data
verifying 27
Invalid characters
eliminating 70
Invalid formats
finding 70
Invalid records
filtering out 73
Invoices
duplicate 69
J
Join command
compare with Relations 50
using 52
M
MAP( ) function
using 70
MATCH( ) function
isolating data 45
Maximum value
MDB files
importing 25
Merchant Category Code list
about 5
Metaphor Corporation
about 3
Methodology
documenting 30
Microsoft Access data
importing 25
Microsoft Excel
data, importing 23, 65
export format 73
Minimum value
Multiple tables
working with 67
K
Key fields
preparing 51, 53, 59
L
Length of field
checking 51, 53, 59, 66
Link
See Hyperlink, command results
Log tab
viewing command entries 15
Logical expressions
filtering data 16
80
N
Navigator pane
about 10
Nested functions
using 70
Notes tab
displaying 11
using 30
Number of records
counting 11
Numerals
using as character data 24
Numeric data
alpha characters in 30
Profile command 11, 12
Statistics command 11
O
Objectives
tutorial 2
Operators
using in filters 38, 45
OR statement
MATCH( ) function 45
Order
ascending or descending 15, 37
columns in the view 10
Order of fields
arranging in view 67
Output fields
selecting 13
Overview of data
obtaining 11
Overview tab
about 10
organizing 64
P
Plain text
export format 35
Policies
credit card usage 4, 5
Profile command
using 12
Project methodology
documenting 30
Project Navigator
location 10
Project planning
Index
analysis strategy 7
Projects, ACL
about 10
adding data 23
creating 22
opening 10
organizing 64
saving 26
viewing content 11
Properties of ACL project
displaying 11
Push-pin button
about 66
Q
Quick Filter
using 55
Quick Sort
about 15
using 37
Quotation marks
character data 36
date data 16
R
Ranges of data
finding 43
Record layout. See Data, format information
Record of commands
Log tab 15
Records
counting 11, 68, 72
extracting 44, 56, 58
invalid
filtering out 73
sorting 37
Relations command
81
compare with Join 50

using 54, 58, 60
Relations dialog box
arranging tables 55
Reports
creating 61
Results
displaying multiple 66
recording 53
saving 37
Reverse quotation marks
dates 16
S
Save
automatic 26
Save the Open Project
button 26
Screen
ACL
described 10
Scripts
writing 66
Search for substrings
using FIND( ) 36
Separators
field 24
Single fields
selecting 11
Sort
Quick 37
Soundness of data
testing 13
Sources
data 6
Spaces
eliminating 45, 70
trimming 43
Start positions
checking 66
Statistics command
using 11
Strings
comparing 70
finding 36
SUBSTR( ). See SUBSTRING( ) function
SUBSTRING( ) function
using 52, 54
Substrings
finding 36
Summarize command
using 42
Summary
tutorial objectives 2
T
Table Layout window
changing data type 35
checking data characteristics 27
displaying 17
Table layouts
comparing 66
editing 28, 65
Tables
adding new columns 52, 54, 58
appending 67
arranging
Relations dialog box 55
combining 50
Company_Dept 57
creating 23
Employees 57
empty 53
exporting 35, 73
extracting records 56, 58
joining 52
Index
methods of combining 67
moving to new folder 64
opening 11
relating 54, 58, 60
removing blank entries 55
saving 26
Unacceptable_Codes 23
Tabs
displaying multiple results 66
Task sheet
sample 30
Test data
using to refine strategy 7
Text
export format 35
Text files
importing 23
Titles of fields
importing 24
Total command
using 15, 34
Total values
Totals
accumulating per value 46
Trans_April worksheet
importing 23
Trans_April.xls
data file 5
Transaction totals
per credit card 46
Transactions
comparing 54
identifying unacceptable 53
isolating by category 43
split 69
summarizing by code 42
Type
drop-down list 35
82
Type of data
changing 24
finding 28
U
Unacceptable_Codes.txt file
about 5
importing 23
Use first row as field names
option 24
V
Validate
compare with verify 70
Validity errors
checking 65
Values
comparing 57
finding in multiple tables 50
Verify
compare with validate 70
Verify button
checking expressions 36
Verify command
using 30, 68
Views
adding new columns 52, 54, 58
column sort order 15
field order
changing 60
removing
blank entries 55
columns 55
W
Web browser
viewing reports 61
Width of fields
adjusting in view 17
Window
Table Layout 17, 35
Worksheets
importing multiple 65
selecting for import 23
2007 ACL Services Ltd.

ACL91 in Practice en

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

ACL91 in Practice en

Caricato da

Copyright:

Formati disponibili

ACL IN PRACTICE

Copyright 2007 ACL Services Ltd.

Check and verify the data ................................................................................................

Chapter 4: Begin Your Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 5: Examine expense patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Chapter 6: Analyze transactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Identify the employees ...................................................................................................... 56

I NTRODUCING M ETAPHOR C ORPORATION

Experiment with real-world data analysis scenarios

ACL in Practice is a tutorial that examines human resources records and

Chapter 1: Introducing Metaphor Corporation

Work with an existing data analysis project

Create and work with a new ACL project from start

Report results of your ACL project

Export the results to another file format

These steps are typical for many data analysis projects

Chapter 2: Examine employee data

Chapter 4: Begin Your Analysis

Determine the total credit limit exposure of the

Determine which credit card numbers are due to

Find which transactions are linked to a particular

Determine some basic information about the

Objective: Examine employee profile data for Metaphor

Examine the project and its contents

Create a statistical overview

Examine employee bonuses and salaries

Chapter 5: Examine expense patterns

Determine transactions per merchant category

Separate the transactions for airfare, hotel stays, or

Determine expenses for drinking establishments

Determine total transaction amounts per credit

Determine transaction amounts per customer

Chapter 3: Set up your project

Create an ACL project.

Import the data into your project. Each data

Verify the data.

Document your work.

About Metaphor Corporation

Chapter 6: Analyze transactions

Determine which transactions are unacceptable,

Create a report that includes employees with

Chapter 7: Validate, correct, and extract data

About Metaphor Corporation

Objective: Expand your analysis to include more data

Import the May 2003 credit card transaction files

Clean up the data in the May 2003 credit card

Extract the repaired data to new tables.

Combine the two May 2003 credit card transaction

Using the techniques you have learned, analyze the

Chapter 1: Introducing Metaphor Corporation

Metaphor credit card policies

By using ACL to analyze the credit card transactions of

Internal office expenses

Metaphors policy is that all purchases must be assigned to

Employees are aware that certain categories of

About Metaphor Corporation

Acceptable category codes

Metaphor data files

Chapter 1: Introducing Metaphor Corporation

Plan your analysis

Plan the analysis

Analyze the data

Reconsider the analysis

View the contents of the employee data project

Generate statistics on the employee list